Protein Modeling

Protein Modeling is a two-part Division C event built around one real protein. Before the tournament, you and your partner study a crystal structure on the computer, then bend a long gray backbone by hand into the same three-dimensional shape and add the side chains and molecules that explain how it works. At the tournament you hand that model in for judging and take a written exam that tests protein chemistry, the biology of the protein's world, and your ability to explore structures in an online viewer called JUDE.

For 2027 the protein is hemagglutinin (HA) from the 2009 H1N1 pandemic influenza virus, and the part you build is the region that grabs sialic acid on the surface of human cells. That choice sets the theme for the whole event. The exam now asks how viruses infect cells and how the immune system fights back, with influenza as the worked case, and at State and Nationals it asks specifically how HA's grip on sialic acid starts an infection. Teams that score well understand one chain of cause and effect: sequence → fold → binding pocket → infection, and they can point to every link on their model.

Event format

The 2027 rules flip the event from last season's design. There is no on-the-spot building from supervisor-provided craft materials and no calculator. You build at home, from a fixed structure, to a fixed scale, and the tournament day is an exam plus a computer-exploration section.

What the rules cover

The 2027 rules manual (Protein Modeling C, Part I and Part II) divides the event into the pre-built model and the written exam. One exam topic is limited to State and National tournaments, and the model's scoring rubric changes at each level.

The material new for 2027 is the protein itself (3UBE, the 2009 pandemic HA), the influenza infection and immune response topic, the State and National sialic-acid topic, and the visualization item on ligands. The rules name a reference article for the build, "Structural Characterization of the Hemagglutinin Receptor Specificity from the 2009 H1N1 Influenza Pandemic" (Journal of Virology, DOI 10.1128/JVI.06322-11), which is free online. Read it: supervisors are likely to draw HA questions from it.

The pre-built model

The target: 3UBE, chain A, residues 153–252

3UBE is a Protein Data Bank entry for an X-ray crystal structure at 2.15 Å of the hemagglutinin from the 2009 pandemic H1N1 virus (strain A/California/04/2009), solved with a human-type receptor analog, the α2,6-linked pentasaccharide LSTc, bound in its pocket. In the file, chain A is HA1 and chain B is HA2; the biological assembly is a trimer of those pairs. You model only 100 residues (153 through 252, inclusive) of chain A. That stretch sits in HA1, the globular head of the protein, and it contains most of the receptor binding site. You do not model the whole trimer, the stem, or other chains in the file.

There are two sources for the coordinates. The 3D Molecular Designs Protein Modeling event page offers a modified file prepared to make modeling easier, and the original atomic coordinates are free from the RCSB Protein Data Bank. Load either into JUDE, restrict the view to the segment you are building, and switch to an alpha-carbon backbone display.

Before you label a residue, check which numbering scheme you are using: the file follows H3 numbering (see the pitfall under The fold of residues 153–252, element by element), while much of the 2009-pandemic literature uses H1 numbering.

Specifications you must meet

Scale math

The alpha-carbon trace replaces each residue with one point, its α-carbon. Neighboring α-carbons are about 3.8 Å apart, because the peptide bond holds them at a fixed distance. At 2 cm per residue, every 3.8 Å of real backbone becomes 2 cm of wire. That single ratio lets you predict the size of every feature before you bend anything.

The fold of residues 153–252, element by element

Before you bend anything, know what you are building. The 100 residues are the receptor-binding subdomain of the HA1 head: almost all β-strands and loops, with one short α-helix. Their sequence, read from the 3UBE coordinates (H3 numbering, one-letter codes, residue 153 first), is:

WLVKKGNSYPKLSKSYINDKGKEVLVLWGIHHPSTSADQQSLYQNADTYVFVCSSRYSKKFKPEIAICPKVRDQEGRMNYYWTLVEPGDKITFEATGNLV

The table lists the secondary-structure elements as the PDB file's own HELIX and SHEET records assign them. Use it as your build plan: mark these boundaries on the wire before you fold. Every distance quoted in the last column is measured in the 3UBE coordinates (the straight line between the element's first and last α-carbon) and converted at 0.53 cm/Å, so these are the lengths a ruler should find on the finished model. They are a little shorter than the textbook rise-per-residue arithmetic predicts, because real helices curve and real strands twist.

Two measurements make the whole build checkable. The α-carbons of residues 153 and 252 are only about 4.4 Å apart (about 2.3 cm on the model): the segment's two ends come back together, because you are modeling a compact subdomain, not a length of chain. And the widest span in the segment is about 45 Å (residues 173 and 189), about 23.5 cm, so the finished backbone occupies roughly a 25 cm box and sits comfortably inside the 50 cm limit.

Across this stretch the two schemes differ by 3 (H1 = H3 − 3), which is why the same change is written D225G in H3 numbering and D222G in H1 numbering. The offset comes from short insertions and deletions earlier in HA1 and is not constant over the whole protein, so convert only residues you have checked, and always say which scheme a legend or an answer uses.

Building it, step by step

1. Map the segment. In JUDE, color by secondary structure and write down, in order from residue 153, where each helix, strand and loop starts and ends. This list is your build plan.
2. Mark the wire. Put a tick at every 2 cm, then label every tenth residue (160, 170, …). Mark the start and end of each helix and strand with tape or a marker.
3. Fold the secondary structure first. Bend helices so the center of the wire traces a circle about 2.4 cm across (a dowel about one wire-thickness narrower helps), keeping 3.6 residues per turn and a right-handed twist. Leave strands nearly straight with a slight zigzag.
4. Fold the tertiary arrangement. Rotate the model on screen until a recognizable feature (a helix, the pocket) faces you, then bend the loops so the helices and strands sit in the same places relative to each other. Check distances: if two α-carbons are 10 Å apart on screen, they should be about 5 cm apart on the model.
5. Add property colors on the α-carbons of the residues you choose to display, using the color code above.
6. Cap the ends. Mark the two ends of the chain so a judge can tell direction at a glance. The convention used in Science Olympiad protein models and in most molecular graphics is blue for the N-terminus and red for the C-terminus (the same blue-to-red "rainbow" that viewers use when they color a chain from start to finish). Put the caps on residue 153 (N-terminal end of your segment) and residue 252, and say in the legend which end is which — that is the one labeling detail that costs nothing and is noticed immediately.
7. Add functional features (side chains, the receptor, labels) with whatever materials you like, including 3D-printed parts.
8. Brace it. Add a stand or internal supports so it survives being lifted and rotated, and check it fits in a 50 cm cube.
9. Write the legend last, so it matches exactly what is on the model.

Functional features and the legend

Every feature you add must explain how the structure relates to function. A feature that is irrelevant, or that you cannot connect to function, earns nothing, so choose few and choose well. For the HA binding site, strong candidates are:

• The pocket floor residues whose side chains form the base of the binding site (aromatic and polar residues that hydrogen-bond to sialic acid).
• The loops and helix that frame the pocket, usually called the 130-loop, 150-loop, 190-helix and 220-loop in H3 numbering. Labeling the ones inside your segment shows judges you know where the site is.
• The receptor itself: a sialic acid, or sialic acid attached to galactose, placed in the pocket with its main contact groups indicated.
• Residues that set receptor preference, such as the aspartates at H3 positions 190 and 225 discussed below.
• Stabilizing pairs at National level: side chains whose hydrogen bonds, salt bridges or hydrophobic contacts hold the pocket's loops in place.

The Model Description Legend is a printed half sheet with a three-column table. For each feature, the first column says what it is, the second says how it is shown on the model, and the third says why it matters. A good row is specific: "Asp190 (H3 numbering) on the 190-helix / red cap on α-carbon plus printed side chain / hydrogen-bonds to the sugars of human-type α2,6 receptors, setting receptor preference." Only residues inside 153–252 have an α-carbon on your model to mark. A weak row says "important residue / red bead / helps binding."

How the model is scored

The 2027 rules give the two parts approximate weights of about 40% for the model and about 60% for the exam, so a team's overall share is roughly $S = 0.40\,M + 0.60\,E$, with M and E the model and exam scores as fractions of their maximums. Treat the weights as approximate: supervisors scale raw points, and the rules do not state an exact formula.

At Regionals the model is judged mainly on how accurately the alpha-carbon backbone is folded: the right helices and strands, at the right scale, packed in the right places. At State and Nationals, accuracy still carries most of the model score, but a growing slice rewards features that explain function and stability. The exam is the bigger lever at every level, so do not spend the last week before a tournament polishing the model and neglecting influenza biology.

Foundations: amino acids and side chains

Proteins are chains of amino acids. The name says what the molecule carries: an amino group (–NH₂, which is –NH₃⁺ at neutral pH) and a carboxylic acid group (–COOH, which is –COO⁻ at neutral pH), both attached to the same carbon, the α-carbon (Cα). The carbon next to the carboxyl group is the α-carbon by definition; the side-chain carbons are then lettered β, γ, δ, ε outward (the amino group at the end of lysine's chain is the ε-amino group). In IUPAC numbering the carboxyl carbon is C1 and the α-carbon is C2, so the standard amino acids are formally 2-amino acids (alanine is 2-aminopropanoic acid); the Greek letters and the IUPAC numbers are two names for the same atoms, offset by one. The fourth bond on Cα goes to the side chain (R group), the only part that differs among the 20 standard amino acids. Its chemistry decides where a residue sits in the fold and what it can do. That is why your model's color code is about side-chain properties, not names.

When amino acids join, each loses the atoms of one water molecule (see Protein calculations); what remains inside the chain is called a residue. A free amino acid has an intact amino and carboxyl group; a residue has given its –OH to one peptide bond and an H to another, so only the chain ends keep a free amine (the N-terminus) and a free carboxylate (the C-terminus). Sequences are always written from N-terminus to C-terminus, the same direction the ribosome builds them.

Some residues sit on the borderline. Tyrosine and tryptophan are aromatic and mostly nonpolar but carry polar groups (–OH and N–H), so different color keys place them differently; say which you chose in the legend. Glycine has only a hydrogen as its side chain and is usually grouped with the hydrophobic residues. Histidine's side chain has a pKₐ near 6, so it is only partly positive at pH 7 and becomes more positive in an acidic endosome, which matters for influenza (see the fusion section).

Textbooks (Lehninger, Berg) use a five-group scheme instead, and most test questions use it, so know both. Nonpolar aliphatic: Gly, Ala, Val, Leu, Ile, Met, Pro. Aromatic: Phe, Tyr, Trp. Polar uncharged: Ser, Thr, Cys, Asn, Gln. Positively charged (basic): Lys, Arg, His. Negatively charged (acidic): Asp, Glu. The only real disagreements between schemes are where to put Gly, Cys, Tyr, Trp, His and Met, so when a question turns on one of those, reason from the side chain's atoms rather than from a memorized list.

Why a side chain is hydrophobic or hydrophilic

Carbon (electronegativity 2.55) and hydrogen (2.20) share electrons almost equally, so a C–H bond is essentially nonpolar. A side chain built only of C and H (Ala, Val, Leu, Ile, Phe) has no partial charges and no N–H or O–H to donate a hydrogen bond, so it cannot hydrogen-bond with water. Water molecules next to it are forced into ordered cages, which costs entropy; burying such groups inside the protein frees that water. This is the hydrophobic effect, covered in Why proteins fold, in energy terms. Side chains with O or N attached to H (Ser, Thr, Tyr, Asn, Gln, His, Lys, Arg, Trp) have polar bonds and can donate and accept hydrogen bonds, and charged side chains (Asp, Glu, Lys, Arg) interact strongly with water and ions. That is why hydrophobic residues (Val, Leu, Ile, Phe, Met) are found mostly in the core of soluble proteins and charged residues (Lys, Arg, Asp, Glu) mostly on the surface.

Two hydrophobicity scales appear on tests. On the Kyte–Doolittle hydropathy scale (positive = hydrophobic), the order is Ile 4.5 > Val 4.2 > Leu 3.8 > Phe 2.8 > Cys 2.5 > Met 1.9 > Ala 1.8 at the hydrophobic end and Arg −4.5 < Lys −3.9 < Asp = Glu = Asn = Gln −3.5 < His −3.2 at the polar end. Among the aromatics, Phe is the most hydrophobic, then Trp, then Tyr (Tyr's –OH makes it the most polar aromatic). Experimental partition scales that measure transfer into octanol rank Trp and Phe as the most hydrophobic of all; when a question asks for "the most hydrophobic amino acid", Ile (Kyte–Doolittle) and Trp or Phe (partition scales) are both defensible, so read the options.

The 20 amino acids at a glance

The table below is the reference you need for "which amino acid…" questions. pKₐ values are Lehninger's textbook values for the free amino acid; different textbooks differ by a few tenths (Asp is often given as 3.9, Cys as 8.3, His as 6.0–6.5), and inside a folded protein the environment can shift them by one or more units. The α-COOH pKₐ is about 1.8–2.4 and the α-NH₃⁺ pKₐ about 8.8–11.0 (Pro highest, 10.96) for every amino acid, so only the side-chain value is listed. Residue mass is the mass of the residue inside a chain (free amino acid minus water, 18.02 Da).

ReferenceThe 20 standard amino acids· 20 rows

Class: NP = nonpolar aliphatic, Ar = aromatic, P = polar uncharged, + = basic, − = acidic. KD = Kyte–Doolittle hydropathy (positive = hydrophobic). Codons = number of codons in the standard genetic code. "—" = no ionizable side chain.

20 rows

Amino acid	3-letter	1-letter	Class	Side chain (key group)	Side-chain pKₐ	pI	KD	Residue mass (Da)	Codons	Notes tests ask about
Glycine	Gly	G	NP	–H	—	5.97	−0.4	57.05	4	Smallest; only achiral one; most flexible backbone; common in turns and every third residue of collagen
Alanine	Ala	A	NP	–CH₃	—	6.01	1.8	71.08	4	Strongest helix former; little steric bulk
Valine	Val	V	NP	isopropyl (branched at Cβ)	—	5.97	4.2	99.13	4	β-branched: favors β-strands; essential
Leucine	Leu	L	NP	isobutyl	—	5.98	3.8	113.16	6	Strong helix former; leucine zipper; essential
Isoleucine	Ile	I	NP	sec-butyl (branched at Cβ)	—	6.02	4.5	113.16	3	Second chiral center; β-branched; most hydrophobic on KD; essential
Methionine	Met	M	NP	–CH₂CH₂–S–CH₃ (thioether)	—	5.74	1.9	131.20	1 (AUG)	Start codon; sulfur cannot form disulfides; "honorary aliphatic"; essential
Proline	Pro	P	NP	ring bonded back to backbone N	—	6.48	−1.6	97.12	4	Secondary amine (no backbone N–H); helix breaker; turns; collagen
Phenylalanine	Phe	F	Ar	benzyl (phenyl ring)	—	5.48	2.8	147.18	2	Most hydrophobic aromatic; essential
Tyrosine	Tyr	Y	Ar	phenol (ring –OH)	10.07	5.66	−1.3	163.18	2	Phosphorylated; absorbs at 280 nm; poorly water-soluble despite –OH
Tryptophan	Trp	W	Ar	indole (benzene fused to pyrrole)	—	5.89	−0.9	186.21	1 (UGG)	Largest; strongest 280 nm absorber; fluorescent; essential
Serine	Ser	S	P	–CH₂OH	—	5.68	−0.8	87.08	6	Phosphorylated; O-glycosylated; catalytic nucleophile in serine proteases
Threonine	Thr	T	P	–CH(OH)CH₃	—	5.87	−0.7	101.10	4	Second chiral center; phosphorylated; O-glycosylated; essential
Cysteine	Cys	C	P	–CH₂SH (thiol)	8.18	5.07	2.5	103.14	2	Disulfide bonds; binds Zn and Fe; R configuration; nucleophile in cysteine proteases
Asparagine	Asn	N	P	–CH₂CONH₂ (amide)	—	5.41	−3.5	114.10	2	N-linked glycosylation site; amide is not basic
Glutamine	Gln	Q	P	–CH₂CH₂CONH₂ (amide)	—	5.65	−3.5	128.13	2	Amide is not basic; nitrogen carrier in metabolism
Lysine	Lys	K	+	–(CH₂)₄NH₃⁺ (ε-amino)	10.53	9.74	−3.9	128.17	2	Acetylated, methylated, ubiquitinated; binds DNA phosphates; essential
Arginine	Arg	R	+	guanidinium	12.48	10.76	−4.5	156.19	6	Most basic; charge spread by resonance; many H-bonds; binds DNA and phosphates
Histidine	His	H	+	imidazole (aromatic ring)	6.00	7.59	−3.2	137.14	2	Only side chain titrating near pH 7; catalytic acid/base; binds Zn, Fe, Ni; essential
Aspartate	Asp	D	−	–CH₂COO⁻	3.65	2.77	−3.5	115.09	2	Salt bridges; metal and Ca²⁺ binding; catalytic acid/base
Glutamate	Glu	E	−	–CH₂CH₂COO⁻	4.25	3.22	−3.5	129.12	2	Salt bridges; good helix former; sickle-cell Glu6

How to read it on a test. Two-pKₐ vs three-pKₐ questions are about the seven residues with an ionizable side chain (Asp, Glu, His, Cys, Tyr, Lys, Arg): they have three pKₐ values (α-COOH, α-NH₃⁺, side chain); the other 13 have two. Codon counts follow from the genetic code: Leu, Ser and Arg have six codons, Met and Trp only one, Ile three, and the 1-letter codes that do not match the first letter are the classic trap (F Phe, Y Tyr, W Trp, N Asn, Q Gln, D Asp, E Glu, K Lys, R Arg).

Residues that get their own questions

A handful of residues behave unlike their class and are tested again and again.

Some general facts about the set, all frequently tested:

• Sulfur: only Cys and Met contain sulfur. Only Cys forms disulfide bonds.
• Rings in the side chain: Phe, Tyr, Trp, His (aromatic) and Pro (non-aromatic). Aromatic heterocycles (ring containing N): His (imidazole) and Trp (indole, which contains a pyrrole ring). Arg has no ring.
• Nitrogen in the side chain: Lys (amine), Arg (guanidinium), His (imidazole), Trp (indole N–H), Asn and Gln (amides). Only Lys, Arg and His are basic.
• Hydroxyl groups: Ser, Thr, Tyr. Small hydroxyl amino acids are very water-soluble because the –OH hydrogen-bonds to water; Tyr is the exception (its large ring makes it poorly soluble).
• Can hydrogen-bond through the side chain: every residue with O or N in its side chain. Gly, Ala, Val, Leu, Ile, Pro, Phe and Met cannot (Met's sulfur is at most a very weak acceptor).
• β-branched (two non-hydrogen groups on Cβ): Val, Ile, Thr. They crowd the backbone in a helix and prefer β-strands.
• Beyond the 20: selenocysteine (the "21st", inserted at a UGA codon with a special mRNA signal) and pyrrolysine (the "22nd", at UAG in some archaea and bacteria). Modified residues such as hydroxyproline and γ-carboxyglutamate are made after translation, not coded.

Essential amino acids, chirality and naming

Essential amino acids are the ones humans cannot synthesize in the amounts needed, so they must come from the diet: His, Ile, Leu, Lys, Met, Phe, Thr, Trp, Val (the classic mnemonic "PVT TIM HALL" lists Phe, Val, Thr, Trp, Ile, Met, His, Arg, Leu, Lys — ten, because older sources count arginine as essential; for 2027 answer nine and leave Arg out). The other 11 (Ala, Arg, Asn, Asp, Cys, Glu, Gln, Gly, Pro, Ser, Tyr) are nonessential, made from metabolic intermediates; several (Arg, Cys, Gln, Gly, Pro, Tyr) are called conditionally essential because the body cannot always make enough (for example, Tyr is made from Phe, Cys from Met's sulfur, and Pro from Glu). Older textbooks sometimes list Arg as essential for children, which is why a count of "10 essential" appears on some tests.

Chirality. Cα of every standard amino acid except glycine carries four different groups, so it is a stereocenter and the amino acid exists as two enantiomers (non-superimposable mirror images), named L and D. Ribosomes build proteins from L-amino acids only; D-amino acids occur in bacterial cell walls and some peptide antibiotics. In the newer R/S system, L-amino acids are S except L-cysteine, which is R, because the sulfur on its side chain outranks the carboxyl group in the priority rules (the shape is the same; only the label changes). Isoleucine and threonine have a second stereocenter in the side chain, so each has $2^n = 2^2 = 4$ stereoisomers (L and D forms of the amino acid, each with allo and normal side-chain forms); in general a molecule with n stereocenters has up to 2ⁿ stereoisomers, and 17 of the 20 standard amino acids have just one stereocenter and two isomers (glycine has none). The handedness of the building blocks is why natural α-helices are right-handed: with L-residues the right-handed helix avoids clashes between side chains and the backbone.

Naming residues and mutations. "Glu1108" or "E1108" means the glutamate at position 1108 of the chain, counted from the N-terminus. A mutation is written old residue, position, new residue: D614G means Asp at position 614 replaced by Gly (the SARS-CoV-2 spike change of 2020), and HA's D222G means Asp to Gly at position 222 in H1 numbering. To judge its effect, compare the two side chains: D614G removes a negative charge and a side chain entirely, so it can break a salt bridge or hydrogen bond and add backbone flexibility.

Charge and pH

The amino and carboxyl groups, and some side chains, gain or lose protons depending on pH. At neutral pH the carboxyl is deprotonated (–COO⁻) and the amino group is protonated (–NH₃⁺), so a free amino acid is a zwitterion: charged groups, zero net charge from the backbone. Side chains with ionizable groups add charge on top of that. Because each amino acid has a group that can donate a proton (–NH₃⁺, –COOH) and a group that can accept one, amino acids are amphoteric: they act as both acids and bases. The carboxyl group is the part most likely to be negative; the amino group is the part that makes an amino acid behave as a base.

The same equation gives the fraction of a group in each form, which is what charge questions really need:

The protonated form of an acidic group is neutral and its deprotonated form is negative; the protonated form of a basic group is positive and its deprotonated form is neutral. So a basic side chain above its pKₐ is, on average, neutral, and an acidic side chain below its pKₐ is neutral.

Net charge of a peptide

A peptide has only one free N-terminus (about +1 below pH ≈ 8) and one free C-terminus (about −1 above pH ≈ 3); the amino and carboxyl groups of the inner residues are tied up in peptide bonds. Add the termini to every ionizable side chain.

For fast multiple-choice screening at pH 7.4, count K and R as +1, D and E as −1, and treat H, C and Y as 0. A sequence such as RFYDW (R +1, D −1) is neutral; FERIW (E −1, R +1) is neutral; RRESRSE has three Arg and two Glu, so it is +1.

Isoelectric point (pI)

The isoelectric point (pI) is the pH at which the molecule's average net charge is zero. It sits midway between the two pKₐ values that flank the neutral form:

Below its pI a protein is net positive; above it, net negative. A protein rich in Asp and Glu has a low pI (acidic protein) and is strongly negative at pH 7; one rich in Lys and Arg has a high pI (basic protein, such as histones, which bind negatively charged DNA). Proteins are least soluble at their pI, because with no net charge the molecules stop repelling each other and aggregate.

Isoelectric focusing (IEF) separates proteins by pI. Proteins move through a gel that holds a stable pH gradient under an electric field; each one migrates until it reaches the pH equal to its pI, where its net charge is zero, and stops. A protein that settles at pH 4 is acidic; one that settles at pH 10 is basic. In 2D gel electrophoresis, the first dimension is IEF (separates by pI) and the second is SDS-PAGE (separates by mass), so each protein becomes a spot. In ordinary electrophoresis at a pH above a protein's pI, the protein is negative and moves toward the positive electrode (anode).

How to read a titration curve: the flat stretches are buffering regions, centered on each pKₐ, where adding base barely changes the pH because the group is absorbing it. The steep jumps are where one group has finished and the next has not started. For glycine, the pI (5.97) lies at the middle of the steep jump between the two flat stretches, at exactly one equivalent of base. A histidine curve buffers near pH 6, which is why histidine is the side chain that responds to pH changes in the physiological range.

pKₐ values shift inside proteins

The table values are for free amino acids in water. Inside a folded protein the local environment can shift a pKₐ by several units. A nearby positive charge or salt bridge stabilizes the negative form of an acid and lowers its pKₐ; burying an Asp or Glu in a nonpolar pocket, where a charge is unfavorable, raises its pKₐ (it stays protonated and neutral); burying a Lys lowers its pKₐ. Hydrogen bonds that stabilize one protonation state have the same effect. This is how active sites tune residues (for example the His of a catalytic triad) and why "standard pKₐ" answers are approximations. Questions that say "ignoring local effects, use the standard pKₐ values" want the table numbers.

Post-translational modifications

A post-translational modification (PTM) is a covalent change made to a protein after (or during) translation. PTMs expand the chemistry of the 20 amino acids: they add charge, bulk, sugars or lipids; switch activity on or off; send a protein to a location; or mark it for destruction. Most are made by enzymes on specific residues, so tests pair each modification with its residue and its effect.

Disulfide bonds in detail. A disulfide forms when two cysteine thiols lose two hydrogens (two protons and two electrons): 2 R–SH → R–S–S–R. That is an oxidation, so disulfides form where conditions are oxidizing: the ER lumen, and therefore in secreted and cell-surface proteins such as HA, insulin, antibodies and keratin. The cytosol is reducing, so cytosolic proteins rarely have them. A reducing agent (dithiothreitol, DTT, or β-mercaptoethanol, BME) adds the hydrogens back and breaks the bond, which is why labs add it to keep cysteines reduced (for example, a zinc finger needs reduced Cys thiols to bind zinc) and why SDS-PAGE run with a reducing agent splits disulfide-linked subunits into separate, smaller bands. A permanent wave in hair works the same way: reduce keratin's disulfides, reshape the hair, then reoxidize to form new ones. The disulfide links residues that may be far apart in sequence, so it stabilizes tertiary structure within a chain, or quaternary structure between chains.

Zymogens. A zymogen (proenzyme) is an inactive precursor that becomes active when a protease cuts off part of the chain. Digestive proteases are made this way (trypsinogen is activated by enteropeptidase in the gut; pepsinogen by stomach acid) so that they do not digest the cells that make them. HA0 is a close analogy: it is not an enzyme, but it must be cut by a host protease before it can work (see Hemagglutinin, from precursor to fusion machine).

Signal peptides and protein targeting

Proteins are sent to the right place by short sequence signals read during or after translation. A protein destined for the ER, the plasma membrane or secretion, such as HA, begins with an N-terminal signal peptide of roughly 15–30 residues with three parts:

The sequence of events: the signal emerges from the ribosome → SRP binds it and pauses translation → SRP docks on its receptor on the ER → the ribosome feeds the chain through the translocon (Sec61) into the ER → signal peptidase removes the signal → the protein folds, forms disulfides and gets its N-linked sugars in the ER, then moves through the Golgi. A second hydrophobic stretch later in the chain becomes a transmembrane anchor; HA has one near its C-terminus, so it ends up in the membrane with its head outside the cell. Other signals: KDEL at the C-terminus keeps proteins in the ER; a short run of Lys and Arg forms a nuclear localization signal; an N-terminal amphipathic helix sends proteins to mitochondria; mannose-6-phosphate tags enzymes for lysosomes.

Protein calculations

Most numeric protein questions reduce to counting residues, bonds and waters, or converting mass to length.

The extra 18.02 Da is the one H on the N-terminus plus the OH on the C-terminus, the only water whose atoms are not lost. The 110 Da rule comes from averaging residue masses weighted by how often each amino acid occurs in proteins. One dalton (Da) is one atomic mass unit, about the mass of a hydrogen atom, and 1 kDa = 1,000 Da (g/mol for molar mass).

Chemistry that drives folding

A protein is a chain that folds itself. The sequence of side chains (Part A) fixes which shapes the chain can reach, the chemistry of the backbone fixes which local shapes are possible at all, and a balance of weak forces in water picks one fold out of an astronomical number. This chapter follows that logic from the bottom up: the peptide bond, the two backbone angles per residue, the regular local shapes (helices, sheets, turns), the recurring motifs and domains built from them, the forces that pack it all into a tertiary and quaternary structure, and the thermodynamics that make folding happen. Every idea here shows up on your model: the 3UBE segment is a stretch of the HA1 head, mostly strands and loops with one short helix (the 190-helix), folded and held by exactly these forces.

Levels of protein structure

Biochemists describe a protein at four levels, each built on the one below. Most "which level" questions are answered by asking which bonds hold that level together.

Some texts add supersecondary structure (motifs such as β-α-β, between 2° and 3°) and domains (independently folding units within 3°), covered below, and a few use quinary structure for the transient contacts proteins make in a crowded cell; quinary is a distractor on most tests.

Hydrogen bonds appear at three levels (2°, 3° and 4°); the peptide bond appears only at the primary level, so a question asking which bond cannot hold a quaternary structure together wants "peptide bond". Changing one residue changes the primary structure by definition and may change everything above it: a deletion such as ΔF508 in the CFTR channel changes 1°, 2° and 3° structure and makes the protein misfold.

The peptide bond and the backbone

Formation. A peptide bond is an amide bond between the α-carboxyl group of one amino acid and the α-amino group of the next. It forms by condensation (dehydration synthesis): the –OH from the carboxyl and an H from the amino group leave as H₂O, joining the carbonyl carbon to the nitrogen (a C–N bond). The reverse, hydrolysis, adds water back. In water, hydrolysis is thermodynamically favored but extremely slow without a catalyst (proteases), and condensation is uphill, which is why cells make peptide bonds on the ribosome using energy stored in aminoacyl-tRNAs. What remains of each amino acid in the chain is a residue (amino acid minus water). Peptide-count and mass arithmetic is in Part A's protein calculations section.

Direction. A chain has a free α-amino group at one end, the N-terminus, and a free α-carboxyl at the other, the C-terminus. Ribosomes add residues to the C-terminus, so proteins are made N → C and sequences are always written N → C. Edman degradation sequences a protein by removing residues one at a time from the N-terminus.

Resonance and planarity. The lone pair on the amide nitrogen is shared with the carbonyl, so the peptide bond is a hybrid of a C–N single bond and a C=N double bond with a negative oxygen and positive nitrogen. Consequences that tests ask about:

• The C–N bond has about 40% double-bond character, so it is shorter than a normal C–N single bond and cannot rotate freely (barrier roughly 80 kJ/mol).
• The six atoms Cα(i), C, O, N, H, Cα(i+1) lie in one plane, the peptide plane (amide plane). Pauling and Corey established this geometry from small-peptide crystal structures and used it to predict the α helix and β sheet in 1951.
• The group is polar: O is partially negative, N–H hydrogen partially positive (dipole about 3.5 debye), which makes the backbone both an H-bond donor (N–H) and acceptor (C=O). Resonance also makes the amide less reactive than an ester or ketone.
• The N is not basic: its lone pair is tied up in resonance, so backbone amides are uncharged at any biological pH.

Bond lengths are a favorite numeric question. The peptide C–N lies between the single and double values, which is the evidence for resonance.

Values differ by about 0.01–0.02 Å between textbooks; learn the ordering (single > peptide > double) and 1.32 Å.

Cis and trans. Because the peptide bond cannot rotate, it is locked as one of two isomers, described by the torsion angle ω: trans (ω ≈ 180°, the two Cα atoms on opposite sides) or cis (ω ≈ 0°, same side). Trans is strongly favored (roughly 1000 : 1 for ordinary residues) because in cis the side chains and Cα atoms of neighboring residues clash. The exception is a bond to proline (X–Pro): proline's ring carbon (Cδ) is bonded to the nitrogen, so both isomers are crowded and the energy gap is small. About 5–6% of X–Pro bonds in folded proteins are cis, versus far below 1% for other bonds; in short unfolded peptides the X–Pro split is much closer to even (roughly 4 : 1 trans : cis). Cis–trans switching at proline is slow (seconds to minutes) and is often the rate-limiting step of folding; enzymes called peptidyl-prolyl cis–trans isomerases (PPIases) catalyze it.

Backbone angles and the Ramachandran plot

Treat the backbone as a chain of flat peptide plates hinged at each α-carbon. Each residue has exactly two hinges:

• φ (phi): rotation about the N–Cα bond.
• ψ (psi): rotation about the Cα–C (carbonyl carbon) bond.
• ω (omega): rotation about the peptide C–N bond, fixed near 180° (trans) or 0° (cis).

A list of (φ, ψ) for every residue specifies the whole backbone conformation, so φ and ψ are the "degrees of freedom" in folding and the quantities Levinthal counted.

The Ramachandran plot (G. N. Ramachandran and colleagues, 1963) plots ψ (vertical) against φ (horizontal), each from −180° to +180°. Most combinations are forbidden by steric clashes: they would push backbone atoms (especially the carbonyl O, amide H and the side-chain Cβ) closer than their van der Waals radii allow. The allowed regions correspond to the secondary structures, which is why the plot gives information about secondary structure.

Positive φ is rare because, for an L-amino acid, it pushes the side-chain Cβ into the preceding carbonyl oxygen. Glycine has no Cβ, so its plot is symmetric and it can use all four quadrants; that is why glycine sits in tight turns and in the αL region. Proline's ring locks φ near −63° (±15°), so proline occupies only a narrow vertical strip (α and PPII regions), and the residue before a proline is also restricted.

How it is tested. Locate α, β and αL on an unlabeled plot; give the quadrant for a left- vs right-handed helix (left-handed: φ ≈ +57°, ψ ≈ +47°, upper right; right-handed: lower left); name which residues can sit in "disallowed" regions (glycine, because it has no side chain; proline is the most restricted); explain why Ramachandran plots are used to validate crystal structures and computer designs (outliers signal steric clashes, i.e., probable model errors). In a well-refined structure about 98% of residues are in favored regions.

Secondary structure

Secondary structure is the regular local shape of the backbone, held by hydrogen bonds between backbone C=O and N–H groups. α helices and β sheets are the two most common kinds; together with turns and loops they account for nearly every residue. They count as secondary structure because they are local, repeating backbone patterns stabilized by backbone hydrogen bonds, not by the overall 3D packing.

The basic chemistry of a hydrogen bond: a hydrogen attached to a strongly electronegative atom (N or O in proteins; F in general chemistry) is attracted to a lone pair on another N or O. In the backbone, the N–H is the donor and the C=O oxygen is the acceptor. Typical donor–acceptor (heavy-atom) distance is about 2.7–3.2 Å, with roughly 3.5 Å the usual cutoff for counting one, and the bond is strongest when N–H···O is close to a straight line. Carbon–hydrogen bonds cannot hydrogen-bond, because carbon (electronegativity 2.55) and hydrogen (2.20) are too similar to make the H partially positive.

The α helix

In an α helix the backbone coils into a right-handed spiral with the side chains pointing outward (and slightly back toward the N-terminus, like branches on a Christmas tree). The C=O of residue i hydrogen-bonds to the N–H of residue i + 4, so every backbone C=O and N–H inside the helix is satisfied; the H-bonds run nearly parallel to the helix axis, all C=O groups point toward the C-terminus and all N–H groups toward the N-terminus. The ring closed by each H-bond contains 13 atoms, hence the formal name 3.6₁₃ helix.

Why right-handed: with L-amino acids, the right-handed coil keeps the Cβ atoms away from the backbone; a left-handed α helix is sterically strained and is found only in short stretches (often with glycine). Mirror-image (D-amino acid) proteins would form left-handed helices.

Helix propensity. Residues differ in how well they fit a helix:

Proline is a strong breaker inside a helix but is common in the first turn: the first three or four N–H groups of any helix have no i − 4 partner anyway, so proline's missing N–H costs nothing there, and its rigid ring helps start the helix. In helix–coil (Zimm–Bragg) theory, the propagation constant s measures how favorable it is to add a residue to an existing helix: s > 1 for strong formers such as Ala, s much less than 1 for Pro.

Charge also matters. A long run of Glu (or Asp) at pH 7 will not form a helix because the side chains are all negative and repel each other; polyglutamate becomes helical at low pH when the carboxylates are protonated. Likewise polylysine is helical only at high pH. Bulky residues packed side by side (a long block of Trp) clash sterically. Pairs of opposite charge at i and i + 4 (or i + 3) sit on the same face, one turn apart, and form helix-stabilizing salt bridges.

Context beats propensity. Helix propensities and the rules above describe an isolated peptide. In a real protein the same stretch of sequence is also feeling long-range (tertiary) interactions: salt bridges to charged residues far away in sequence, packing against a hydrophobic core, metal or ligand binding, and contacts with another subunit. Those can easily override the local preference, in either direction. A run of glutamates that would repel itself into a coil on its own becomes helical when each carboxylate is paired with a lysine or arginine from elsewhere in the fold, or when the pH is lowered; a sequence of strong helix formers is pulled into a strand or a loop when the fold demands it; and the same sequence can adopt different secondary structures in different proteins (so-called chameleon sequences). This is why secondary-structure prediction tops out near 80–85% accuracy even with evolutionary profiles (single-sequence methods are closer to 70%), and why a design question that says "make this segment helical" is usually asking you to add a stabilizing partner somewhere else in the chain, not just to swap in alanines.

The helix dipole. Each peptide group is a small dipole, and in a helix they all point the same way. The sum gives the helix a macrodipole, with a partial positive end at the N-terminus (where the N–H groups point) and a partial negative end at the C-terminus (where the C=O groups point); the effect is often described as about half a unit charge at each end. Helices are therefore stabilized by negative residues (Asp, Glu) near the N-terminus and positive residues (Lys, Arg, His) near the C-terminus, and helix N-termini often point at phosphate or carboxylate groups of bound ligands.

Amphipathic helices and the helical wheel. Because successive residues are 100° apart, residues at i, i + 3, i + 4, i + 7, i + 11 land on the same face. A helix with hydrophobic residues on one face and polar or charged residues on the other is amphipathic: the hydrophobic face packs against the protein core or a membrane and the polar face looks at water. A helical wheel projects the residues onto a circle viewed down the axis so you can see the faces.

Other helices: 3₁₀, π, polyproline II and collagen

Helices are named n_m: n residues per turn, m atoms in the ring closed by each hydrogen bond.

How common is each? In globular proteins roughly 30–35% of residues sit in α helices and 20–25% in β strands, with most of the rest in turns and loops. Among the minor helices the ranking is: 3₁₀ is by far the most common (it caps the ends of many α helices, though only about 3–4% of residues), polyproline II next (a few percent, in proline-rich surface segments and unfolded chains), and the π helix rarest (short bulges). When a description fits both 3₁₀ and PPII — 3.0 residues per turn, 120° per residue — the separating facts are that a 3₁₀ helix is right-handed with internal i→i+3 hydrogen bonds, while PPII is left-handed with no internal hydrogen bonds; and if the question asks which is more common, it is the 3₁₀ helix.

Why the α helix wins. Its hydrogen bonds are nearly linear and its backbone atoms pack against each other in van der Waals contact all along the axis: an α helix has no hollow core, which is the single most common misreading of a cartoon or ribbon drawing (the ribbon is drawn through the chain, not around a tube of empty space). The 3₁₀ helix is wound too tightly, so its i→i+3 hydrogen bonds are bent and its side chains crowd each other; the π helix is wound too loosely, leaving a hole down the middle that is too small for water but too large to pack, so it costs van der Waals energy. Both therefore appear only in short stretches, usually as the first or last turn of an α helix or as a bulge at a functional site.

Helix formation is also cooperative. Making the first i→i+4 hydrogen bond requires freezing a whole turn of chain, about three residues, with no hydrogen-bond payback yet, so nucleation is expensive; after that, each added residue completes one more hydrogen bond at little extra entropy cost, so propagation is cheap and the helix "zips" onward. Two consequences tests ask about: helices form and melt in an all-or-none way rather than residue by residue, and a short isolated peptide (fewer than about 15–20 residues) is mostly random coil in water, because it can make too few hydrogen bonds to pay for nucleation and has no hydrophobic core to bury. Helices in real proteins are stabilized by packing against the rest of the fold.

Collagen is the classic fibrous helix. Its sequence repeats Gly–X–Y, with X often proline and Y often 4-hydroxyproline (Hyp). Glycine must be every third residue because that position faces the crowded center of the triple helix, where only a hydrogen fits; a single Gly substitution causes brittle-bone disease (osteogenesis imperfecta). Hydroxylation of proline needs vitamin C; without it collagen is unstable (scurvy). Collagen is not an α helix: test-makers love "triple helix" as the answer.

β strands and β sheets

A β strand is a nearly fully extended stretch of backbone (about 3.2–3.5 Å per residue, so a 6-residue strand is about 20 Å long) in which successive side chains point to alternate sides. A strand alone has no hydrogen bonds. A β sheet forms when two or more strands lie side by side and hydrogen-bond between strands, backbone C=O to backbone N–H, with the H-bonds roughly perpendicular to the strands. The sheet is pleated because each α-carbon is tetrahedral; side chains stick up and down from the pleats, never in the plane of the sheet. Strands sit about 4.8 Å apart.

Sheets can also be mixed (some neighbors parallel, some antiparallel). Real sheets are twisted, not flat; a strongly twisted sheet can close into a β barrel. β-strand formers are the β-branched and aromatic residues Val, Ile, Thr, Tyr, Phe, Trp; an alternating pattern such as Val–Ile–Val–Ile is a textbook strand sequence, and an alternating hydrophobic/polar pattern (one face to the core, one to water) marks an edge or surface sheet. Inside a sheet the strands that have only one neighbor are edge strands; they are where sheets add partners (and where amyloid can grow).

Sheet topology is written as the order of strands across the sheet, numbered by their order in the sequence.

Silk fibroin is the classic fibrous β protein: stacked antiparallel sheets rich in Gly–Ala repeats (small side chains let the sheets pack tightly). It is strong and flexible but does not stretch, because the chains are already fully extended. Compare α-keratin (hair, nails), whose α helices can be stretched; heating wet hair breaks hydrogen bonds so it can be reshaped, and a permanent wave breaks and remakes disulfide cross-links (see the tertiary section).

Turns, hairpins and loops

Helices and strands are roughly straight, so a globular chain needs ways to reverse direction. About a third of residues in a globular protein are in turns and loops, almost always on the surface.

The β turn (also called a reverse turn or β bend) reverses the chain in four residues. Type I and type II differ by a 180° flip of the central peptide plane. Proline is favored at position i + 1 (its fixed φ suits the turn) and glycine at i + 2 in type II (the positive φ there suits glycine best). Asn, Asp and Ser are also common in turns. A cis proline at i + 2 gives a type VI turn.

A β hairpin is two adjacent antiparallel strands joined by a short turn (2–5 residues); an Asn–Gly pair (or D-Pro–Gly in designed peptides) at the turn strongly favors a tight two-residue hairpin turn; ordinary L-Pro–Gly does not, because it makes a type II turn whose twist does not fit the strands. Hairpins are the simplest β motif and have low contact order (the partners are close in sequence), so they form fast.

Loops are irregular connections longer than a turn, with no repeating φ/ψ. "Random coil" in a structure file does not mean random or disordered: a loop has a defined conformation in the folded protein, it is just not regular. Loops are rich in polar residues, Gly and Pro, tolerate insertions and mutations, and often carry function: antibody epitopes and binding sites are usually loops. In HA, the 130-loop, 150-loop and 220-loop form the rim of the receptor binding site together with the 190-helix, so many functional residues on your model sit on loops (see the receptor binding site section).

Predicting secondary structure from sequence

Tests ask you to look at a sequence and guess its structure. The rules of thumb (the logic of the Chou–Fasman method):

• Runs rich in Ala, Leu, Glu, Met, Gln, Lys, Arg, with hydrophobic residues spaced 3–4 apart → α helix.
• Runs of Val, Ile, Tyr, Phe, Trp, Thr, especially alternating hydrophobic/polar → β strand.
• Pro, Gly, Asn, Asp, Ser clusters → turn or loop; a Pro in the middle of a would-be helix ends or kinks it.
• About 20 consecutive hydrophobic residues → a transmembrane helix (see protein classes below).

Experimentally, circular dichroism (CD) in the far ultraviolet (about 190–250 nm) measures overall secondary-structure content: α helices give negative bands at about 208 and 222 nm, β sheets a single negative band near 218 nm, unfolded chains a negative band near 200 nm.

Motifs, domains and folds

Between secondary and tertiary structure are recurring combinations of helices and strands. Three words get mixed up:

• A motif (or supersecondary structure) is a small, recurring arrangement of a few secondary-structure elements (β-α-β, β hairpin, helix-turn-helix), or a short conserved sequence pattern with a function (such as the NPA motif of aquaporin water channels or the RGD cell-attachment motif). A motif usually does not fold on its own.
• A domain is a compact region of one chain that folds independently and often has its own function, typically about 50–250 residues. Large proteins are built from several domains; the same domain type (an SH2 domain that binds phosphotyrosine, an SH3 domain that binds proline-rich sequences, a kinase domain, a DNA-binding domain) recurs in many proteins.
• A fold is the overall topology (which elements, in what order and arrangement) shared by domains that may have unrelated sequences.

HA illustrates all three. The HA1 head (where your segment lies) contains the receptor-binding domain, built mainly on an eight-stranded antiparallel β sandwich of the jelly-roll type, decorated by the loops and short helix that form the binding pocket. The HA2 stem is dominated by long α helices that pack into a coiled coil at the trimer's center. Structural similarity with low sequence identity is common: HAs of different subtypes share a fold despite large sequence differences, which is evidence that structure is more conserved than sequence.

Supersecondary motifs and common folds

Structural biologists classify domains into four broad classes: all-α (myoglobin, four-helix bundles), all-β (immunoglobulins, jelly rolls), α/β (helices and mostly parallel strands alternating along the chain: TIM barrel, Rossmann) and α+β (helical and strand regions in separate parts of the chain).

Contact order is the average sequence separation between residues that touch in the native structure, divided by chain length. Local structures (helices, hairpins) have low contact order; a sheet that pairs the N-terminal and C-terminal strands has a high one. Small proteins with low contact order fold faster.

DNA-binding and metal-binding motifs

Many tests from earlier seasons drilled transcription-factor motifs. They are general protein chemistry and still fair game.

Residues that contact DNA are mostly positively charged (Lys, Arg), which also attract the negative phosphate backbone, and H-bonding polar residues (Asn, Gln, Ser, Thr); base-specific contacts come from H-bonds and van der Waals contacts with the edges of the bases in the major groove. Nonpolar residues such as Ile are poor at reading bases. Arg reading guanine and Asn/Gln reading adenine are classic pairs.

The C2H2 zinc finger in detail. A finger is about 30 residues with the consensus pattern Cys–X₂₋₄–Cys–X₁₂–His–X₃₋₅–His. The two cysteines come from the β hairpin and the two histidines from the C-terminal part of the helix; together they hold one Zn²⁺ in a tetrahedral site (Zn–S about 2.3 Å, Zn–N about 2.0–2.1 Å). The zinc is structural: without it the 30-residue peptide cannot fold, because it is too small to have a real hydrophobic core. It does not touch the DNA (its positive charge may help offset the phosphates slightly). Side chains at a few positions on the helix's DNA-facing surface (classically positions −1, 3 and 6 relative to the helix start) make the base contacts, so a mutation there abolishes recognition even though the finger still folds. Other zinc-finger fold groups include the treble clef, zinc ribbon, gag knuckle (retroviral nucleocapsid), C4 fingers of nuclear receptors and GATA factors, and the RING finger. About 2–3% of human genes encode C2H2 zinc-finger proteins.

Why zinc? Zn²⁺ is redox-inactive (it never changes oxidation state), so it can cross-link a small domain inside the reducing cytosol, where disulfide bonds would not survive. The site must stay reduced: researchers add reducing agents (DTT, β-mercaptoethanol) to zinc-finger samples so the cysteine thiols stay free to bind zinc rather than forming disulfides. In experiments, Co²⁺ (or Cd²⁺) can replace Zn²⁺ in the same site, which is how the tetrahedral geometry was confirmed spectroscopically. Zinc deficiency or mutation of a coordinating Cys/His leads to misfolding and loss of DNA binding. Metal binding can be measured by atomic absorption spectroscopy.

Protein classes: globular, fibrous and membrane

Proteins fall into three broad structural classes, set by their environment and job. Most proteins are globular: the large majority of the soluble proteins in a cell (enzymes, transport and signaling proteins, antibodies) fold into compact, roughly spherical shapes, while fibrous and membrane proteins are smaller groups with special jobs.

A globular protein in water folds inside-out relative to water: nonpolar residues (Val, Leu, Ile, Phe, Met, Trp) in the core, charged and polar residues (Lys, Arg, Asp, Glu, Asn, Gln, Ser) on the surface where they stay hydrated. Aromatic residues are mostly in the core. Tau stays unfolded largely because it is so hydrophilic that no hydrophobic collapse drives it.

Membrane protein topology. In a membrane, the "solvent" around the middle of the protein is the bilayer's hydrocarbon core, about 30 Å thick. So membrane proteins put hydrophobic side chains on the outside of the transmembrane part and hydrophilic residues on the loops that face the water on either side.

• α-helical transmembrane segments are about 20 hydrophobic residues (20 × 1.5 Å ≈ 30 Å, just spanning the core). In a helix every backbone C=O and N–H is H-bonded internally, which matters because there is no water to satisfy them in the lipid. Hydropathy plots (Kyte–Doolittle, averaged over windows of about 19 residues) find these segments as peaks of high hydrophobicity.
• GPCRs (rhodopsin, adrenergic receptors) cross the membrane seven times (7 TM helices). The interior of a helix bundle can be polar, forming a pocket for polar ligands or a channel for ions or water; the hydrophobic faces point out at the lipid.
• β-barrel membrane proteins (porins in the outer membranes of bacteria, mitochondria and chloroplasts) use strands whose side chains alternate: hydrophobic facing out into lipid, polar facing the water-filled pore.
• Positively charged residues cluster on the cytoplasmic side (the "positive-inside" rule), and Trp/Tyr often sit at the water–lipid interface.
• HA is a type I single-pass membrane protein: one transmembrane helix near the C-terminus of HA2, the large N-terminal ectodomain (head and stem) outside the virus, and a short cytoplasmic tail inside.

Tertiary and quaternary structure

Tertiary structure is the full 3D fold of one chain; quaternary structure is the assembly of several chains. HA is a trimer of three identical HA1–HA2 units, so the complete protein has quaternary structure, but your 100-residue segment shows tertiary structure within one chain. Hemoglobin (α₂β₂ tetramer) is the textbook quaternary example. Subunits let a protein be built from small, error-tolerant pieces, allow cooperativity and allostery (Part C), and let fibrous proteins combine individually weak strands into a strong cable (two α-keratin helices coil around each other).

The same forces build both levels. The table gives rough energies and distances; the energies are per interaction in water and vary a lot with environment.

Hydrogen bonds and water. Inside a folded protein, a hydrogen bond between two groups replaces hydrogen bonds each group made with water, so the net gain from forming it is small. That is why hydrogen bonds set the specific shape (secondary structure, binding) but the hydrophobic effect provides most of the driving force. In a nonpolar environment (a membrane interior, or a solvent such as hexane with dielectric constant about 2 versus about 80 for water) backbone hydrogen bonds no longer compete with water and electrostatics are stronger, so helices form more readily and are more regular. Leaving an N–H or C=O unpaired in the core is costly.

Salt bridges and buried charges. Electrostatic interactions follow Coulomb's law.

Charges therefore interact far more strongly in the low-dielectric interior. But a buried charge first has to give up its water (a large desolvation cost), so an isolated charged residue in the core is strongly destabilizing; it is tolerated only when paired with an opposite charge or H-bond partner. The protein environment can also shift side-chain pKₐ values (a nearby salt bridge or H-bond stabilizes one protonation state; see Part A, Charge and pH).

Disulfide bonds. Two cysteine thiols (–SH) are oxidized to a disulfide (–S–S–), the pair being called cystine:

Only cysteine forms disulfides; methionine's sulfur is a thioether with no free –SH. Disulfides form in the oxidizing endoplasmic reticulum (catalyzed by protein disulfide isomerase, PDI) and are found in secreted and cell-surface proteins. The cytosol is reducing (high glutathione), so disulfides are rare inside cells; intracellular proteins use metal sites or simply packing instead. Disulfides stabilize mainly by lowering the entropy of the unfolded chain (the loop they tie cannot spread out). They can link two parts of one chain (tertiary) or two chains (quaternary): HA1 and HA2 stay attached after HA0 cleavage through an interchain disulfide, and HA has several intrachain disulfides formed in the ER. In SDS gel electrophoresis, adding a reducing agent breaks disulfides so disulfide-linked chains run separately.

Hair, a worked application. α-keratin is rich in cysteine. A permanent wave reduces the disulfides, the hair is curled, and an oxidant re-forms disulfides in new positions. Heat straightening works on hydrogen bonds instead: too little heat leaves the H-bonds intact and the hair keeps its shape.

Why proteins fold, in energy terms

Folding happens because the folded (native) state has lower Gibbs free energy than the unfolded one under physiological conditions. ΔG of folding is negative; equivalently, ΔG of unfolding is positive.

The entropy balance. An unfolded chain can take a vast number of conformations, so folding lowers the chain's entropy (ΔS_chain < 0), an unfavorable cost. Folding also forms many hydrogen bonds, salt bridges and van der Waals contacts (ΔH contributions, partly canceled by the water bonds they replace). The decisive term comes from the solvent: water surrounding exposed nonpolar groups forms ordered, cage-like (clathrate-like) shells with fewer allowed hydrogen-bonding arrangements. When the chain buries those groups, the water is released and the entropy of the water increases a lot. So folding is driven by a decrease in the entropy of the protein but an increase in the entropy of water as hydrophobic residues are buried. That entropy gain is the hydrophobic effect, the largest driving force for folding, and the same effect drives membrane and vesicle formation. Forming water cages is the cost, not the benefit.

Marginal stability. The net ΔG of unfolding of a typical small protein is only about 20–60 kJ/mol (roughly 5–15 kcal/mol), the difference between two huge, nearly canceling terms and equal to just a few hydrogen bonds. That is why one mutation, a few degrees of heat, extreme pH or urea can tip a protein over, and why "native proteins are thousands of times more stable than unfolded ones" is false in energy terms (though even 20 kJ/mol gives an equilibrium of thousands to one, as below).

Melting temperature and the stability curve. The melting temperature T_m is where ΔG of unfolding = 0, K = 1 and half the molecules are unfolded (the midpoint of the transition; above T_m the unfolded state is favored). Because unfolding exposes nonpolar groups to water, the unfolded state has a larger heat capacity (ΔC_p > 0), which makes ΔG(T) a downward-curving parabola: stability peaks near room temperature and falls on both sides. That is why some proteins also unfold when cooled (cold denaturation): at low temperature water solvates nonpolar groups better and the hydrophobic effect weakens.

What changes stability. A stabilizing mutation lowers the free energy of the folded state relative to the unfolded state: by adding favorable contacts in the native state (better core packing, an H-bond or salt bridge) or by lowering the entropy of the unfolded state (Gly → Ala, adding Pro in a loop, adding a disulfide). Stability (equilibrium, ΔG) and folding speed (kinetics, the activation barrier) are separate: a mutation can change one without the other. Destabilizing changes include putting a charged or proline residue in the hydrophobic core (core Ile → Pro is about the worst single change), removing a buried hydrophobic group (leaving a cavity), or breaking a disulfide. Solution conditions matter too: pH (protonation states), ionic strength (salt screens electrostatics), and attached glycans (which can stabilize HA and other glycoproteins) all affect stability.

Anfinsen, Levinthal and the folding funnel

Anfinsen's experiment (ribonuclease A, 1950s–60s; Nobel Prize 1972). RNase A has 124 residues and 4 disulfide bonds. Anfinsen unfolded it with 8 M urea (disrupts non-covalent interactions) plus β-mercaptoethanol (reduces the disulfides). Activity vanished.

• Remove both by dialysis (urea out, then air oxidation re-forms disulfides in the folded chain): activity returns almost completely and the native disulfides re-form.
• Remove the reducing agent first, so disulfides form while urea keeps the chain unfolded: the 8 cysteines pair randomly, giving "scrambled" RNase with about 1% activity. Adding a trace of β-mercaptoethanol lets disulfides reshuffle and the native form slowly returns.

Conclusion (Anfinsen's dogma / thermodynamic hypothesis): the amino acid sequence contains all the information needed for the native fold, and the native state is the thermodynamically most stable (lowest free energy) state under physiological conditions. The native structure is the most populated state at equilibrium even if other local minima exist.

Stated carefully, Anfinsen's dogma has conditions attached, and questions often ask for them:

When any condition fails, the dogma's prediction fails with it: large multi-domain proteins need chaperones to stay kinetically accessible, proteins such as HA are metastable rather than at the global minimum, and prions show that a sequence can occupy a second, stable fold.

Levinthal's paradox (Cyrus Levinthal, 1969). If a chain had to try every conformation, folding could never finish. With 3 states per residue, a 100-residue chain has 3¹⁰⁰ ≈ 5 × 10⁴⁷ conformations; sampling one every 10⁻¹³ s would take about 5 × 10³⁴ s ≈ 10²⁷ years, far longer than the age of the universe (about 1.4 × 10¹⁰ years). Real proteins fold in microseconds to seconds (minutes when proline isomerization is involved). The resolution: folding is not a random search; local interactions form quickly and bias the search, so the chain moves downhill on a guided path. (Different sources count differently: 3 states for each of the 2 angles per residue gives 3²⁰⁰, and Levinthal's own estimate was about 10³⁰⁰. The method, multiply the states per unit together, is what is tested.)

The folding funnel. Modern theory pictures an energy landscape shaped like a funnel: the wide top is the huge number of high-energy unfolded conformations, the narrow bottom is the native state. Any chain moves downhill by many routes; the width of the funnel at each height represents conformational entropy. Rough patches and side pockets are local minima (kinetic traps) where partly folded or misfolded molecules can get stuck, which is when chaperones help. The transition state is the highest-free-energy point (a saddle) that molecules must pass to reach the native state. The principle of minimal frustration (Bryngelson and Wolynes) says evolution has selected sequences whose native contacts are mutually compatible, giving a smooth, funnelled landscape so folding is both stable and fast; random sequences are frustrated and fold poorly.

Folding pathways. Folding usually passes through intermediates:

• Hydrophobic collapse: nonpolar residues quickly drive the chain into a compact state, inside which secondary and tertiary structure then form.
• Framework (diffusion–collision) model: local secondary structure forms first and the pieces then dock.
• Nucleation–condensation: a small native-like nucleus forms and the rest condenses around it.
• The molten globule is a common intermediate: compact, with much native-like secondary structure, but with loose, fluid side-chain packing (no fixed tertiary structure).

Folding starts co-translationally: the N-terminal part of a chain can fold while the rest is still being made on the ribosome. That, plus chaperones and folding enzymes, is why some proteins fold in cells in seconds but take days (or fail) in a test tube, and why proteins expressed in another species sometimes misfold.

Kinetic versus thermodynamic control: metastable proteins

Anfinsen's hypothesis says the native state is the global free-energy minimum (thermodynamic control). Some functional proteins are instead kinetically trapped in a metastable native state: a lower-energy state exists, but a high barrier keeps them out of it until a trigger arrives (kinetic control).

Influenza HA is the event's own example. After HA0 is cleaved, the pre-fusion HA trimer is metastable, "spring-loaded". At endosomal pH (or, in the lab, with heat or urea at neutral pH) it crosses the barrier and refolds into the more stable post-fusion form, in which a loop of HA2 becomes helical and extends the central coiled coil, firing the fusion peptide toward the target membrane (see the hemagglutinin section). The energy released by that one-way refolding does the work of membrane fusion. Serpins (protease inhibitors) are another metastable class. Prions show the dark side: PrP^Sc is a stable, non-native fold of the same sequence, often cited as an exception to Anfinsen's dogma.

Denaturation and denaturing agents

Denaturation is the loss of native 2°, 3° and 4° structure without breaking peptide bonds, so the primary structure survives. Denatured proteins usually lose function. Whether denaturation is reversible depends on the protein: small single-domain proteins (RNase A is the classic case) often refold when conditions return; large, multidomain or aggregated proteins (cooked egg white) do not.

SDS in SDS-PAGE does three jobs: it denatures proteins, coats them with uniform negative charge per unit mass, and forces them into rod-like shapes, so they separate by size alone. A prolonged fever is a mild everyday example: sustained high temperature begins to denature heat-sensitive proteins.

Measuring folding. Circular dichroism (secondary structure content), tryptophan fluorescence (buried Trp emits near 330 nm; exposed Trp shifts toward 350 nm), NMR, X-ray crystallography, differential scanning calorimetry (T_m and ΔH), and hydrogen–deuterium exchange: in D₂O, backbone amide hydrogens exchange for deuterium (mass increases; NMR peaks vanish), and the H-bonded, buried amides of the folded core exchange slowest, mapping which parts are protected.

Chaperones and folding helpers

Cells do not change the thermodynamics of folding; they change the odds of getting there. Molecular chaperones bind exposed hydrophobic segments of nascent or partly folded chains, prevent aggregation and misfolding, and release the chain (usually with ATP hydrolysis) for another attempt. They associate with their clients only while those clients are folding and do not alter the final native structure or the folding equilibrium. Many are heat-shock proteins (Hsps), made in larger amounts when heat increases unfolding.

The difference tests ask about: a chaperone binds directly to an exposed segment; a chaperonin encloses the protein. HA folds in the ER: its disulfides form there (with PDI), calnexin and calreticulin bind its N-linked glycans, BiP binds early forms, and only correctly folded monomers assemble into trimers and leave for the Golgi. Misfolded proteins are retained and sent for degradation (ubiquitin–proteasome), which is also the fate of the misfolded ΔF508 CFTR in cystic fibrosis.

Misfolding, amyloid and prions

A non-native state is any conformation of the chain other than its functional native fold. It can be an unfolded or partly folded intermediate, or a misfolded state: a compact but wrong structure, often trapped in a local free-energy minimum, that the protein reaches from the same sequence in its normal environment (as opposed to unfolding caused by a hostile environment such as heat or denaturant). When chains misfold, exposed hydrophobic surfaces can stick together. Many misfolded proteins convert to amyloid: long fibrils with a cross-β structure, β strands running perpendicular to the fibril axis and hydrogen bonds running along it. Amyloid is extremely stable and resists proteases.

How this chapter is tested. Expect definitions of the four levels and which bonds hold each; peptide-bond chemistry (condensation, resonance, bond length, trans versus cis, why rigid); φ/ψ and Ramachandran quadrants; helix numbers (3.6, 1.5 Å, 5.4 Å, i + 4, 3₁₀ = i + 3, π = i + 5), turn residue counts, helix breakers and formers, the helix dipole; parallel versus antiparallel sheets; motif identification (zinc finger ββα with Cys₂His₂, leucine zipper = coiled coil, Greek key = 4 antiparallel strands, flavodoxin = sheet between helices); which interaction is covalent, strongest or dominant; disulfide oxidation and reduction; the sign and size of ΔG; the hydrophobic effect as water entropy; Anfinsen, Levinthal, funnel, molten globule; denaturant mechanisms; chaperone versus chaperonin; and prions.

Protein function and binding

Every question in this chapter comes back to one sentence: a protein's three-dimensional shape puts specific side chains in specific places, and those side chains decide what it binds and what it does. Learn the vocabulary below precisely, because many exam items are definitions.

The table covers the words; the rest of the chapter explains the mechanisms behind them. Notice that "active site" is a special case of "binding site": every enzyme has a binding site, but a binding site (like HA's) need not catalyze anything.

What proteins do

Protein functions are usually sorted into a handful of classes. Tests ask you to match a protein to its class or to explain how its structure suits the job, so connect each class to a shape.

Many proteins belong to more than one class. HA, for example, is a receptor-binding protein, a membrane-fusion machine and, because antibodies target it, the main antigen of the virus (its "three functions" are receptor binding, fusion and antigenic variation; see Hemagglutinin, from precursor to fusion machine).

Specificity comes from complementarity

A ligand binds where its shape and its chemistry match the site: a hydrophobic ring meets a hydrophobic patch, a carboxylate meets an arginine or a backbone N–H, a hydroxyl meets a hydrogen-bond partner. Each contact is weak (a few kJ/mol), so only a ligand that makes many contacts at once binds well; a molecule that is almost right misses some and falls off. Because the site is built from L-amino acids, it is chiral: a drug's mirror image (enantiomer) usually fits poorly, which is why many drugs are sold as a single enantiomer.

Binding changes function in four general ways, and a free-response answer should name the one that applies:

1. Blocking: the ligand occupies a site another molecule needs (competitive inhibitors; antibodies covering HA's receptor pocket).
2. Stabilizing a conformation: the protein exists in several shapes, and the ligand holds it in one (allosteric activators and inhibitors; O₂ shifting hemoglobin toward the R state).
3. Triggering a shape change that is passed on (receptors relaying a signal; low pH triggering HA's fusion rearrangement).
4. Recruiting partners: the bound protein now has a new surface for other proteins (phosphotyrosine recognized by SH2 domains).

Enzymes: how catalysis works

An enzyme is a biological catalyst, almost always a protein. James Sumner showed in 1926 that enzymes are proteins by crystallizing urease (Nobel Prize 1946), settling a long dispute about what enzymes were made of.

Activation energy (Eₐ or ΔG‡) is the energy needed to reach the transition state, the highest-energy, least-stable arrangement of atoms partway between reactants and products, in which old bonds are partly broken and new ones partly formed. The transition state exists for about the length of one bond vibration and cannot be isolated. The higher the barrier, the smaller the fraction of molecules with enough energy to cross it per second, and the slower the reaction.

An enzyme binds the transition state more tightly than it binds the substrate. Holding the substrate in a strained, transition-state-like shape and pre-arranging the right charges around it lowers the energy of the transition state, so Eₐ drops and the rate rises, often by factors of a million or more. What an enzyme does not do is equally testable:

• It does not change ΔG of the reaction, the equilibrium constant, or the final ratio of products to reactants. It only gets there faster.
• It speeds the forward and reverse reactions by the same factor.
• It is not consumed, so a small amount converts many substrate molecules.
• It does not make an unfavorable (ΔG > 0) reaction happen by itself; cells do that by coupling to ATP hydrolysis.

Catalytic strategies

Enzymes combine a few chemical tricks. Exam answers that name the trick and the residue that performs it score well.

A nucleophile is an electron-rich atom that donates an electron pair to an electron-poor atom (an electrophile) to form a bond. Cysteine's thiol (–SH, readily ionized to –S⁻) is the strongest nucleophile among the side chains, which is why cysteine shows up in many active sites; serine's –OH becomes a strong nucleophile only when a neighboring histidine pulls its proton away (the catalytic triad).

Lock and key versus induced fit

The lock-and-key model (Emil Fischer, 1894) pictures a rigid active site that already matches the substrate. The induced-fit model (Daniel Koshland, 1958) says the site changes shape as the substrate binds, closing around it and moving catalytic groups into position. Induced fit is closer to reality for most enzymes: it explains why some molecules bind without reacting (they bind but fail to induce the catalytic shape) and how an enzyme can exclude water from the site. A related idea, conformational selection, says the protein already samples several shapes and the ligand binds and stabilizes the matching one.

Cofactors

Many enzymes need a non-protein helper. The protein part alone is the apoenzyme; with its cofactor it is the active holoenzyme.

The table separates cofactors by how they are held, not by what they do. Zinc is a good example of one ion with two roles: in carbonic anhydrase it is catalytic, while in a zinc finger it is purely structural (it holds the fold together). Removing a structural metal unfolds the motif, which is why zinc deficiency can impair zinc-finger proteins.

Enzyme classes and names

Enzyme names usually end in -ase and describe the reaction. The Enzyme Commission (EC) sorts all enzymes into seven classes.

Proteases are worth extra detail because influenza depends on them. Serine proteases (trypsin, chymotrypsin, thrombin, the airway protease TMPRSS2) use a Ser–His–Asp triad; cysteine proteases (papain, caspases, cathepsins) use a Cys–His pair and cut proteins; aspartic proteases (pepsin, HIV protease) use two Asp; metalloproteases use Zn²⁺. Trypsin-like proteases cut after Lys or Arg, which is how host proteases cleave HA0 at its arginine to activate it. Many digestive proteases are made as inactive zymogens (trypsinogen, pepsinogen) and activated by cleavage so they do not digest the cell that makes them (see Post-translational modifications).

Factors that change enzyme activity

Questions often list a change in conditions and ask how activity responds. Reason from structure: anything that changes the active site's shape or charge changes activity.

Notice which factors act by denaturing (high temperature, extreme pH) and which act reversibly without unfolding (substrate level, cooling, reversible inhibitors). A modest temperature change (say 10 K within the normal range) changes the rate but usually not the binding of the natural substrate, whereas a pH change that alters a catalytic histidine's charge can abolish activity. For which interactions each denaturant breaks, see Why proteins fold, in energy terms.

Enzyme inhibition

An inhibitor is a molecule that lowers an enzyme's activity. Inhibitors are classified by where they bind and whether they can come off. This is the single most tested enzyme topic, and the key distinction is whether adding more substrate rescues activity.

The first three rows are reversible: the inhibitor binds by non-covalent forces and comes off again. Read the table by columns: "where it binds" tells you the type, and "does substrate overcome it" is the experiment that distinguishes them. The kinetic column is qualitative only: the algebra of Michaelis–Menten kinetics is outside the 2027 scope, but the vocabulary is not, and the table above uses it.

Allosteric regulation and feedback

Allostery (Greek "other shape") means a molecule binding at one site changes the protein's shape and activity at another site. Allosteric enzymes are usually multi-subunit and switch between a low-activity T (tense) state and a high-activity R (relaxed) state. An allosteric inhibitor stabilizes T; an allosteric activator stabilizes R.

Cells use allostery for feedback inhibition: the end product of a pathway inhibits the enzyme at the first committed step, so the pathway shuts off when product is plentiful. Classic example: CTP inhibits aspartate transcarbamoylase, the first enzyme of pyrimidine synthesis, while ATP activates it. Covalent modification is the other major control: phosphorylation by kinases adds a negatively charged phosphate (about −2 at pH 7) that can form new salt bridges with Lys and Arg and flip the protein into a different conformation; phosphatases reverse it (details in Post-translational modifications). Kinases that drive the cell cycle, such as cyclin-dependent kinase 1 (CDK1), are switched on by binding cyclin and by phosphorylation, an example of both mechanisms at once.

Hemoglobin and myoglobin: binding without catalysis

Hemoglobin and myoglobin are the textbook pair for reversible binding, cooperativity and allostery. Questions come up every season, and the same ideas explain how any multi-subunit binding protein, HA included, can behave differently from a single chain.

The table's last three rows all follow from one fact: hemoglobin's subunits talk to each other and myoglobin has no partners. Myoglobin was the first protein structure solved by X-ray crystallography (John Kendrew, 1958), and hemoglobin followed (Max Perutz, 1959); they shared the 1962 Nobel Prize in Chemistry.

How O₂ binds. Each heme is a flat porphyrin ring holding an Fe²⁺ ion. The iron is bonded to four ring nitrogens and, below the ring, to the proximal histidine (His F8); O₂ binds on the other side, where the distal histidine (His E7) hydrogen-bonds to it. Only ferrous iron (Fe²⁺) binds O₂; oxidized methemoglobin (Fe³⁺) cannot. Carbon monoxide binds the same iron about 200–250 times more tightly than O₂, which is why CO is poisonous; the distal His hydrogen-bonds bound O₂ (but not CO), which keeps that preference far lower than it is for free heme in solution.

Cooperativity. In deoxyhemoglobin (T state) the iron sits slightly out of the ring plane. When O₂ binds, the iron moves into the plane and pulls His F8 and its whole F helix with it; that movement shifts the α₁β₂ contacts and nudges the tetramer toward the high-affinity R state. So the first O₂ binds weakly, but each one makes the next easier, giving the S-shaped curve. The practical payoff: hemoglobin is nearly saturated in the lungs (about 98% at 100 torr) yet unloads a large fraction in tissues, far more than a hyperbolic binder with the same P₅₀ could. Two textbook models describe this: the concerted (MWC) model, in which all subunits switch T→R together, and the sequential (KNF) model, in which subunits change one at a time.

Allosteric regulators all shift the curve to the right (lower affinity, more O₂ released):

• H⁺ and CO₂: the Bohr effect. In active tissue, CO₂ and lactic acid lower the pH. Protonated His residues form salt bridges that stabilize the T state; the classic one pairs His146 of the β chain with Asp94 of the same chain. Protonated His (pKₐ about 6–7) is positive and attracts the negative Asp; at higher pH, His loses its proton and the salt bridge breaks. CO₂ also binds the N-termini as carbamate, adding more T-state salt bridges.
• 2,3-Bisphosphoglycerate (2,3-BPG) binds the positively charged central cavity between the two β chains in the T state only (contacts include β His2, Lys82 and His143). Fetal hemoglobin (α₂γ₂) has Ser instead of His at position 143, binds BPG weakly, and so has higher O₂ affinity than the mother's, which lets O₂ cross the placenta.
• Temperature: warmer tissue also shifts the curve right.

Sickle-cell disease shows how one substitution changes binding behavior: Glu6 of the β chain becomes Val (written E6V), putting a hydrophobic patch on the surface that sticks into a pocket on a neighboring deoxyhemoglobin, so deoxy-HbS polymerizes into fibers (see Mutations and sequence changes).

Receptors, channels and antibodies

Receptors bind a signal molecule and change shape to pass the message on. The two membrane receptor families that tests compare:

The transmembrane helices of a GPCR face lipid on the outside, so their outer surfaces are hydrophobic (Leu, Ile, Val, Phe, Ala); polar residues line the inside of the bundle, forming the pocket where usually polar ligands bind. Proline is uncommon in the middle of transmembrane helices because it kinks them, although a few conserved kinks are functional. A third family, receptor tyrosine kinases (insulin receptor, EGF receptor), dimerize on binding and phosphorylate each other's tyrosines; the phosphotyrosines then recruit SH2-domain proteins.

Membrane transport proteins follow the same topology rule: a band of hydrophobic surface where the protein crosses the bilayer, hydrophilic surfaces in the water on either side. Channels open a pore and let ions flow down their gradient (fast); carriers bind a solute and change shape to release it on the other side (slower); pumps use ATP to move ions against their gradient. Influenza's M2 is a proton channel (see Influenza virus).

Antibodies are the defense proteins most relevant to influenza. An IgG antibody is a Y-shaped, roughly 150 kDa protein of two heavy and two light chains joined by disulfide bonds. The two tips (Fab arms) each carry six hypervariable loops (complementarity-determining regions, CDRs) that form the binding surface, the paratope, which fits a patch on the antigen, the epitope. The stem (Fc) recruits immune cells and complement. Antibodies that cover HA's receptor pocket block attachment (neutralization); the immune-system side is in Adaptive immunity.

Proteins that bind DNA

Protein–DNA recognition is a favorite source of binding questions. DNA is a polyanion: every phosphate carries a negative charge. Proteins that bind it therefore have positively charged surfaces rich in Lys and Arg, which make non-specific salt bridges with the backbone phosphates (histone tails use exactly this).

Sequence-specific recognition happens mostly in the major groove, where the edges of the base pairs expose a distinctive pattern of hydrogen-bond donors, acceptors and methyl groups for each pair (A·T, T·A, G·C and C·G all look different there). An α-helix fits neatly into the major groove, so most DNA-binding motifs place a "recognition helix" there. Typical side-chain contacts: Arg makes two hydrogen bonds to guanine, Asn and Gln pair with adenine, and hydrophobic residues contact the methyl group of thymine. Small residues such as Ser and Ala, or nonpolar ones such as Ile, give little base-specific information. The minor groove is narrower and less distinctive.

The common motifs (helix–turn–helix, the C2H2 zinc finger with its ββα fold and Zn²⁺ held by two Cys and two His, the leucine zipper, helix–loop–helix) are described under DNA-binding and metal-binding motifs. For binding questions, remember four facts about classic C2H2 zinc fingers:

• The α-helix sits in the major groove, and side chains at a few positions near its start contact about three base pairs per finger.
• Fingers are strung together in tandem; each adds about three base pairs, so several fingers give a long, highly specific site, and deleting one finger weakens binding.
• The Zn²⁺ is structural: without it the finger unfolds and stops binding DNA.
• About 2–3% of human genes (roughly 700–800) encode C2H2 zinc-finger proteins, the largest family of transcription factors.

DNA methylation at CpG sites adds a methyl group to cytosine (5-methylcytosine), and that methyl projects into the major groove. It can block a transcription factor that needs the unmethylated base, or create a site for methyl-CpG-binding repressor proteins (including some zinc-finger proteins), which recruit machinery that silences the gene. A base change outside a protein's binding site usually does not affect binding. A protein can still bind and yet fail to activate a gene if the change prevents it from recruiting partner proteins (co-activators or co-repressors). Histone acetylation neutralizes Lys charges, loosens chromatin and gives transcription factors access to DNA.

How function questions are asked

• Definitions: ligand, active site, binding pocket, activation energy, transition state, cofactor, zymogen, allostery, cooperativity.
• "What does an enzyme change?": Eₐ, yes; ΔG, Keq and products, no.
• Inhibitor identification from a description: "resembles the substrate" means competitive; "binds elsewhere and changes the shape" means allosteric/noncompetitive; "binds only after substrate" means uncompetitive; "covalent" means irreversible.
• Predict the effect of adding substrate to an inhibited enzyme.
• Name a residue for a catalytic role: His (acid–base), Ser or Cys (nucleophile), Asp or Glu (acid, metal binding), Lys or Arg (binds phosphate or DNA).
• Hemoglobin vs myoglobin: curve shapes, cooperativity, Bohr effect, BPG, fetal Hb, CO.
• Receptor comparisons: GPCR (7 TM, slow, amplified) versus ion channel (fast).
• Connect to HA: HA binds (it does not catalyze); NA catalyzes; oseltamivir is a competitive transition-state analog of NA; antibodies neutralize by blocking HA's pocket.

Mutations and sequence changes

The central dogma

Francis Crick stated the central dogma in 1958: sequence information passes from nucleic acid to nucleic acid and from nucleic acid to protein, but never from protein back to nucleic acid. In ordinary cells that means DNA is copied (replication), DNA is transcribed into RNA (transcription), and RNA is translated into protein (translation). A protein's amino-acid sequence (primary structure) is therefore set by the DNA sequence of its gene, and the primary structure in turn determines how it folds.

The special transfers matter for viruses. Retroviruses (HIV) use reverse transcriptase to copy their RNA genome into DNA, so their information path is RNA → DNA → RNA → protein; reverse transcription is the usual answer to "which process seems to violate the central dogma?" (Crick's version actually allowed it; the textbook one-way version does not). Influenza is a negative-sense RNA virus: its RNA-dependent RNA polymerase copies the genome into positive-sense mRNA and into new genomes, so its path is RNA → RNA → protein with no DNA at all (see The infection cycle). Prions are sometimes mentioned as an exception, but they pass on a misfolded shape, not a sequence, so they do not break the dogma.

Transcription and RNA processing

Transcription copies one strand of a gene into RNA. RNA polymerase binds a promoter, unwinds the DNA and reads the template (antisense) strand from 3′ to 5′, adding ribonucleotides to the new RNA's 3′ end, so the RNA grows 5′→3′ and is antiparallel to the template. Base pairing is G with C, and A in the template pairs with U in the RNA (RNA has uracil instead of thymine). The other DNA strand, the coding (sense) strand, has the same sequence as the mRNA except T in place of U.

In eukaryotes the primary transcript (pre-mRNA) is processed in the nucleus before translation: a 5′ cap (7-methylguanosine) is added, the 3′ end gets a poly(A) tail, and introns are spliced out so that exons join. Proteins are therefore not translated from unprocessed mRNA in eukaryotes. Alternative splicing joins different exon combinations, so one gene can make several protein isoforms. The exome is the set of all exons, the protein-coding part of the genome (only about 1–2% of human DNA). Bacteria have no nucleus, so their ribosomes start translating mRNA while it is still being transcribed.

Translation and the genetic code

Translation happens on ribosomes, in the cytoplasm or on the rough endoplasmic reticulum (for secreted and membrane proteins such as HA). Each codon of three mRNA nucleotides specifies one amino acid or a stop signal. A tRNA carries an amino acid at one end and an anticodon at the other that base-pairs antiparallel with the codon; aminoacyl-tRNA synthetases attach the correct amino acid to each tRNA. The ribosome links each new amino acid to the growing chain, so the protein grows from its N-terminus to its C-terminus as the mRNA is read 5′→3′.

Use the codon table below for any translation question. It is laid out by first base (blocks of four rows), then third base, so the four codons of a four-fold degenerate family sit on the same row.

ReferenceThe standard genetic code (all 64 codons)· 16 rows

Codons are mRNA triplets read 5′→3′. For DNA, replace U with T. Read a row to see how the third base usually matters least (wobble). Reverse lookup: the amino acid column tells you every codon that encodes a residue.

16 rows

Codon	Amino acid	Codon	Amino acid	Codon	Amino acid	Codon	Amino acid
UUU	Phe (F)	UCU	Ser (S)	UAU	Tyr (Y)	UGU	Cys (C)
UUC	Phe (F)	UCC	Ser (S)	UAC	Tyr (Y)	UGC	Cys (C)
UUA	Leu (L)	UCA	Ser (S)	UAA	STOP (ochre)	UGA	STOP (opal)
UUG	Leu (L)	UCG	Ser (S)	UAG	STOP (amber)	UGG	Trp (W)
CUU	Leu (L)	CCU	Pro (P)	CAU	His (H)	CGU	Arg (R)
CUC	Leu (L)	CCC	Pro (P)	CAC	His (H)	CGC	Arg (R)
CUA	Leu (L)	CCA	Pro (P)	CAA	Gln (Q)	CGA	Arg (R)
CUG	Leu (L)	CCG	Pro (P)	CAG	Gln (Q)	CGG	Arg (R)
AUU	Ile (I)	ACU	Thr (T)	AAU	Asn (N)	AGU	Ser (S)
AUC	Ile (I)	ACC	Thr (T)	AAC	Asn (N)	AGC	Ser (S)
AUA	Ile (I)	ACA	Thr (T)	AAA	Lys (K)	AGA	Arg (R)
AUG	Met (M) — START	ACG	Thr (T)	AAG	Lys (K)	AGG	Arg (R)
GUU	Val (V)	GCU	Ala (A)	GAU	Asp (D)	GGU	Gly (G)
GUC	Val (V)	GCC	Ala (A)	GAC	Asp (D)	GGC	Gly (G)
GUA	Val (V)	GCA	Ala (A)	GAA	Glu (E)	GGA	Gly (G)
GUG	Val (V)	GCG	Ala (A)	GAG	Glu (E)	GGG	Gly (G)

Two habits make codon questions fast: look at the first two bases first, because they usually fix the amino acid, and check the strand's direction before reading anything. The table is your reference for codon questions; the worked example shows the method. Folding begins co-translationally: the N-terminal part of a chain can start folding while the rest is still being made, and chaperones assist.

Kinds of mutations

A mutation is a heritable change in DNA sequence. Classify it at the DNA level and then by its effect on the protein.

Nonsense and frameshift mutations are the two classic ways to knock out a gene. A frameshift inside an intron is usually silent, because the intron is spliced out before translation (unless it hits a splice signal). Mutations are written as original residue, position, new residue: E6V (Glu6→Val, sickle cell), D614G (Asp614→Gly in the SARS-CoV-2 spike), ΔF508 (Phe508 deleted in CFTR).

Some mutation outcomes have names worth knowing. A loss-of-function mutation removes activity; a gain-of-function mutation adds or increases one. A dominant negative mutation makes a defective subunit that still assembles with normal ones and spoils the whole complex, so one bad copy is enough to block function. Multi-subunit proteins are vulnerable to this: a mutant subunit of a dimeric transcription factor that still dimerizes but cannot bind DNA, or an HA monomer that joins a trimer but cannot fold correctly, would poison the assembly.

Predicting the effect of a substitution

How to rank possible replacements for a residue: match polarity and charge first, then size and shape, and treat anything nonstandard as out of the running. For Thr, the best substitute is Ser (same –OH, one methyl smaller), then Val (same β-branched shape, but loses the –OH) or Asn; a charged or bulky aromatic residue ranks last. A substitution matrix such as BLOSUM62 encodes the same logic from how often each swap is seen between related proteins: Ile↔Leu↔Val, Ser↔Thr, Asp↔Glu, Lys↔Arg and Phe↔Tyr score highest.

Ask three questions about any amino-acid change: Where is the residue (surface, core, binding site, interface)? What changes (size, charge, polarity, hydrogen bonding, backbone flexibility)? What contact is lost or gained?

A conservative substitution swaps an amino acid for one with similar size and chemistry, and usually has a small effect. Common conservative pairs: Lys↔Arg (both positive), Asp↔Glu (both negative), Leu↔Ile↔Val↔Met (hydrophobic aliphatic), Phe↔Tyr↔Trp (aromatic), Ser↔Thr (small hydroxyl), Asn↔Gln (amides), Asp↔Asn. A change that looks small in shape can still be non-conservative in chemistry: Thr→Val keeps the shape (Val has CH₃ where Thr has OH) but removes a hydroxyl that hydrogen-bonds, so it is polar→nonpolar; Asp→Gly (D614G) removes a charge and a whole side chain and makes the backbone more flexible.

Stability in energy terms. A stabilizing mutation makes the folded state lower in free energy relative to the unfolded state, so ΔG of unfolding becomes larger and positive (more energy needed to unfold) and the melting temperature rises. Two routes: add favorable contacts to the folded state (better packing, a new salt bridge or disulfide), or reduce the entropy of the unfolded state (for example a Gly→Ala or X→Pro change in a loop, or a disulfide, both of which restrict the unfolded chain). A destabilizing mutation does the reverse (see Why proteins fold, in energy terms).

Disease examples tests reuse

Three diseases recur as worked cases of single changes with large structural consequences. Each one illustrates a different row of the location table above: a surface change that creates a sticky patch, a deletion that blocks folding, and changes in a DNA-binding site.

In each case the answer that earns credit names the level of structure affected: sickle cell changes how folded tetramers pack together (quaternary contacts between molecules), ΔF508 changes folding itself, and p53 hot-spot mutations remove DNA contacts or destabilize the domain that makes them.

Influenza: mutation in action

Influenza's RNA-dependent RNA polymerase has no proofreading, so it makes roughly one error per 10⁴–10⁵ nucleotides copied, about one new mutation per genome per replication. Most are harmful or neutral, but selection keeps the useful ones:

• Antigenic drift: substitutions in the antigenic sites around the HA head change the surface antibodies recognize, so last season's antibodies bind poorly.
• Receptor-specificity changes: substitutions in the pocket, such as at H3-numbered positions 190 and 225 (D222G in H1 numbering), shift which sialic-acid linkage the virus prefers (see What sets receptor preference in H1 HAs).
• Cleavage-site changes: inserting several basic residues at HA0's cleavage site lets furin-like proteases throughout the body activate HA, a hallmark of highly pathogenic avian strains.
• Drug resistance: the H275Y substitution in neuraminidase (N1 numbering) makes it resistant to oseltamivir.

Antigenic shift, the swap of a whole genome segment by reassortment, is a different process (see Influenza virus).

Conservation and sequence alignment

A residue is conserved if it stays the same (or nearly the same) at the matching position across related proteins or species. A sequence alignment lines up two or more sequences, inserting gaps where needed, so that matching positions sit in the same column; tools such as BLAST (for searching databases) and Clustal (for multiple alignments) do this. Identity is the percent of aligned positions that are the same amino acid; similarity also counts conservative substitutions. Homologous proteins share a common ancestor (an all-or-nothing statement); percent identity is the evidence for it.

Conservation is evidence of importance. Natural selection removes mutations that damage function, so residues that do essential jobs (catalytic residues, metal ligands, glycines in tight turns, prolines at kinks, disulfide cysteines, receptor-pocket residues) change least, while surface loops change most. In HA, the receptor-pocket floor residues (Tyr98, Trp153, His183, Tyr195 in H3 numbering) are conserved across subtypes, while the antigenic sites around them vary from year to year. Proteins with low sequence identity can still share the same fold, because structure is more conserved than sequence: as few as 25–30% identical residues usually means the same overall fold.

How mutation questions are asked

• Transcribe and translate a template or coding strand; find a start codon; spot a stop.
• Name the mutation type from a DNA change or from the protein outcome (truncated → nonsense or frameshift).
• Pick the most or least disruptive substitution (look for charge into core, Pro into helix, loss of Cys or catalytic residue; Lys→Arg, Leu→Ile, Asp→Glu are mild).
• Explain why a conserved residue matters, or why active-site residues are conserved.
• Central dogma statements, retroviral exceptions, and true/false items on wobble, introns and processing.
• Influenza links: drift, receptor-switching substitutions, cleavage site, resistance.

Viruses and how they infect cells

What a virus is, and what it is not

A virus is a set of genes packed in a protein shell (the capsid), sometimes wrapped in a lipid envelope. It has no metabolism of its own, so it must enter a host cell and use the cell's machinery — ribosomes, tRNAs, amino acids, ATP, membranes — to copy itself. Sizes run roughly 20–300 nm, one to two orders of magnitude smaller than a bacterium, and genomes run from about 2 kb to a few hundred kb.

What a virus lacks is the usual exam point: no ribosomes, no plasma membrane of its own making, no cytoplasm, no ATP synthesis, no growth, no binary fission. Virions do not get bigger and then split; they are assembled from parts. Outside a cell a virion is inert particles — which is why viruses are usually described as replicating rather than reproducing, and why "is a virus alive?" has no settled answer.

Some vocabulary that tests use interchangeably and should not:

Building a shell out of many copies of one small protein is forced on viruses by genome economy: a capsid large enough to hold the genome must be made from repeated subunits, because coding one protein per capsid position would need far more DNA than the capsid can hold. That constraint produces the two common symmetries — icosahedral (20 triangular faces, quasi-spherical) and helical (subunits spiralling around the nucleic acid). Influenza is a hybrid: helical nucleocapsids inside a roughly spherical (often pleomorphic or filamentous) envelope.

Enveloped versus naked viruses

The envelope is the single structural fact that predicts the most about a virus's behavior, so questions about it are common.

Soap works on influenza for a purely chemical reason worth stating: soap and alcohol dissolve the lipid envelope, so the spikes fall apart and the particle cannot attach or fuse. Naked viruses have no lipid to dissolve, which is why norovirus outbreaks need bleach rather than hand sanitizer.

The Baltimore classification

David Baltimore's scheme sorts viruses by the chemical nature of the genome and the number of steps between that genome and mRNA, because every virus must make mRNA the host ribosome can read. Learn it as "how many steps to (+) mRNA?"

Two consequences the bank asks about directly. First, a (+) ssRNA virus can be infectious as naked RNA, because injecting the genome alone is enough to start translation; a (−) ssRNA virus cannot, because the cell has no enzyme that reads RNA from RNA, so the polymerase must come packaged in the virion. Second, Class VI reverses the usual DNA → RNA direction of the central dogma: reverse transcriptase makes DNA from an RNA template, which is the exception every test cites (see Mutations and sequence changes for the dogma itself). Information flow for a retrovirus is RNA → DNA → RNA → protein.

The generic infection cycle

Every viral infection follows the same broad sequence, and exam questions often ask you to place an event in it:

1. Attachment (adsorption): a viral surface protein binds a specific molecule (the receptor) on the host cell. Receptor choice sets which species and tissues the virus can infect (its tropism).
2. Entry (penetration): the virus enters by fusing its envelope with the plasma membrane, or by being taken up in an endosome and escaping from it; naked viruses breach the endosome membrane or form a pore.
3. Uncoating: the genome is released from the capsid.
4. Replication and expression: viral genes are copied and translated by host ribosomes. Early proteins are usually enzymes; late proteins are usually structural.
5. Assembly (maturation): new genomes and proteins come together, often with a proteolytic cleavage step that activates the particle.
6. Release: by budding (enveloped viruses take their envelope from the host membrane) or by bursting the cell.

The interval between uncoating and the first new virions is the eclipse period, when no complete virus can be recovered from the cell — a classic "what is happening at point X on the one-step growth curve?" question. Burst size is the number of virions released per infected cell (hundreds to thousands for influenza).

Virologists count virus with a handful of standard measures, and a growth-curve or titer question assumes them:

On a one-step growth curve (high MOI, so every cell is infected at the same moment), the titer first falls or stays flat during the eclipse period, then rises steeply as progeny are released, then plateaus; the rise's height divided by the number of cells is the burst size.

Entry routes worth naming: direct fusion at the plasma membrane (HIV, pH-independent), receptor-mediated endocytosis followed by low-pH fusion (influenza), endosomal escape without fusion (adenovirus), and injection of the genome only (bacteriophages, which leave the capsid outside).

Lytic and lysogenic cycles

These two words come from bacteriophage biology but appear on protein-modeling tests as general virus vocabulary.

Influenza has no lysogenic option: it is an acute, lytic-style infection with no integration step, because its genome is RNA and it encodes no integrase. Lysogeny also explains transduction, in which a phage carries host genes between bacteria, and why some bacterial toxins (diphtheria, cholera) are encoded on prophages.

Influenza virus

Structure of the virion

Influenza A is an enveloped, negative-sense single-stranded RNA virus (family Orthomyxoviridae). Its genome is split into eight segments, each wrapped with nucleoprotein (NP) and bound to its own polymerase heterotrimer (PB1, PB2 and PA) to form a ribonucleoprotein (vRNP), which folds back on itself as a double-helical hairpin. Under the envelope lies a layer of matrix protein (M1), the particle's structural scaffold. Three proteins sit in the envelope:

Particles are pleomorphic: laboratory-adapted strains are roughly spherical at about 100 nm, while clinical isolates are often long filaments micrometers in length. That variability is why electron-microscope questions ask about "roughly spherical to filamentous" rather than a fixed shape, and why influenza cannot be given a crystallographic symmetry the way an icosahedral virus can.

A full strain name is written in a fixed order: type / host (omitted if human) / place of isolation / strain number / year (subtype). The 2027 protein comes from A/California/04/2009 (H1N1) — type A, human, isolated in California, isolate 04, in 2009, subtype H1N1. Tests love asking you to decode such a name or to say what is missing (the host, because it is human).

Influenza types A, B, C and D

The pattern to remember: only A has the animal reservoir wide enough and the subtype diversity needed for antigenic shift, so only influenza A causes pandemics. C and D get away with seven segments because HEF merges the jobs of HA and NA — and because their receptor is not Neu5Ac but 9-O-acetylated sialic acid, which the esterase removes.

The eight segments and what they encode

The segments are numbered by decreasing length. Every one of them must be packaged for a virion to be infectious, which is why co-infection so readily produces reassortants.

Two structural tricks are on this list. Segments 7 and 8 each make two proteins by alternative splicing of the viral mRNA in the nucleus — an unusual move for an RNA virus and one of the reasons influenza must replicate there, since splicing happens only in the nucleus. Segments 2 and 3 each make an extra protein from a different reading frame of the same RNA. Both are genome-economy devices: eight segments, far more than eight proteins.

The infection cycle

1. Attachment. HA on the virion binds sialic acid at the tips of glycans on airway epithelial cells. Each contact is weak; many HA–sialic acid contacts at once make the attachment stick (avidity, below).
2. Endocytosis. The cell takes the bound virus into an endosome, mostly by clathrin-mediated endocytosis.
3. Acidification. The endosome pumps in protons and its pH falls to about 5–6. M2 lets protons into the virion interior, which loosens M1 from the vRNPs so they can be released after fusion.
4. Fusion. Low pH (about 5.0–5.5 for most human strains) triggers a large, irreversible change in HA's shape that drives the viral and endosomal membranes together (details below). The vRNPs are released into the cytoplasm.
5. Nuclear import and replication. Unlike most RNA viruses, influenza copies its genome in the nucleus; NP's nuclear localization signals get the vRNPs through the nuclear pore. The viral polymerase makes mRNA, using 5′ caps snatched from host mRNAs to prime transcription (cap-snatching: PB2 grabs the cap, the PA endonuclease cuts 10–13 nucleotides downstream, and PB1 extends that fragment). Transcription makes capped, polyadenylated mRNA; replication instead makes a full-length positive-sense copy, the cRNA, which serves as template for new (−) vRNA.
6. Translation and assembly. HA, NA and M2 are translated on ER-bound ribosomes, glycosylated, folded and sent through the Golgi to the apical plasma membrane; M1, NP, NS1 and the polymerase subunits are made on free ribosomes. New vRNPs leave the nucleus with M1 and NEP and gather under patches of membrane rich in viral proteins.
7. Budding and release. New virions bud from the apical surface, taking host membrane as their envelope, with M2 helping to pinch off the particle. HA on the new virions would stick them right back onto sialic acid on the same cell, so NA cuts sialic acid to set them free.

The viral RdRp has no proofreading exonuclease, so it misincorporates roughly one nucleotide in $10^{4}$–$10^{5}$ — about the length of a single segment. Every infected cell therefore produces a cloud of slightly different genomes (a quasispecies), which is the raw material for antigenic drift and for drug resistance.

Hemagglutinin, from precursor to fusion machine

HA is made as one chain, HA0, which folds and assembles into a homotrimer in the ER. A host protease must cut HA0 into HA1 and HA2, held together by a disulfide bond, before the virus can infect — HA0 binds sialic acid perfectly well but cannot fuse. The cut is therefore an activation step exactly like a zymogen's (see Post-translational modifications), and uncleaved particles are non-infectious.

• HA1 forms the globular head on top. It carries the receptor binding site and most of the antibody targets. Your model segment, chain A residues 153–252, is part of HA1.
• HA2 forms most of the long stem (the triple-helical coiled coil) and anchors the trimer in the membrane. It holds the fusion peptide, a hydrophobic stretch of about 20 residues at the new N-terminus created by cleavage, buried in a charged pocket in the pre-fusion shape. Its sequence in H1 HAs begins GLFGAIAGFIE…: glycines and alanines for flexibility and small size, two phenylalanines and two isoleucines for bulk hydrophobicity, and almost no polar residues until the glutamate, which is exactly the composition that lets it bury itself in the hydrocarbon core of the target membrane. It is a favorite side-chain-properties question because every letter in it can be justified from the chemistry.

At endosomal pH the HA1 heads loosen their grip on each other, and HA2 refolds in what is the textbook example of a spring-loaded conformational change: a loop between two helices becomes helical, extending one long helix that throws the fusion peptide roughly 100 Å toward the endosomal membrane, where it inserts. The molecule then jackknifes, folding the C-terminal part of HA2 back alongside the long helix into a hairpin, which drags the viral membrane (held by the transmembrane anchor) against the endosomal membrane (held by the fusion peptide). Several trimers acting together bend the two bilayers into a hemifusion stalk and then a pore.

The trigger is protonation: as pH falls, histidine side chains (pKa ≈ 6) and carboxylates pick up protons, salt bridges that clamped the pre-fusion state break, and the molecule falls into the lower-energy post-fusion shape. The change is irreversible — the pre-fusion state is metastable, a loaded spring, and the energy released does the work of fusion. That is also why HA exposed to low pH outside a cell is permanently inactivated and non-infectious, and why acid-stability of HA is studied as a transmissibility trait.

Antigenic drift, shift and pandemic history

Drift is mutation within a subtype (H3N2 of 2015 versus H3N2 of 2025); shift changes the subtype itself (H1N1 → H2N2). A mixing vessel host — classically the pig, whose airway carries both avian-type and human-type receptors — allows an avian and a human virus to meet in one cell and swap segments.

Notice the pattern in 1957 and 1968: the new virus kept some segments, so people with antibodies to the retained N2 were partly protected, and those pandemics were milder than 1918. Note too that H1N1 disappeared from human circulation after 1957, reappeared in 1977, and was then replaced by the 2009 virus — so "H1N1" on a test can mean three different viruses, and the year matters.

The 2009 H1N1 pandemic

The 2009 pandemic virus arose in pigs by reassortment, the swapping of whole genome segments between two influenza viruses that infect the same cell. Its segments came from two swine influenza lineages, which themselves carried genes of swine, avian and human origin — which is why it is called a quadruple reassortant:

Its HA is an H1 from the classical swine lineage, which descends from the 1918 pandemic virus. That ancestry explains a striking feature of 2009: many older adults, exposed decades earlier to 1918-like H1 viruses, had antibodies that cross-reacted with the new HA, while children and young adults did not. The age distribution of severe cases was therefore inverted relative to a normal flu season — one of the clearest real-world demonstrations of immune imprinting.

The immune response to influenza

Innate immunity

The first defenses act within minutes to hours and do not depend on prior exposure. They are non-specific (they respond to classes of molecule, not to one strain), germline-encoded (not rearranged, so no new specificities can be made), and they have no memory — a second infection gets the same innate response as the first.

Barriers come first. The airway is lined by a mucus blanket moved by beating cilia (the mucociliary escalator), which traps virions and sweeps them to the throat to be swallowed. Mucus is itself rich in heavily sialylated glycoproteins (mucins), so it acts as a decoy receptor: HA sticks to mucus instead of to cells, and NA must cut the virus free. Lysozyme, defensins, surfactant proteins A and D (which bind HA's glycans), low skin pH and stomach acid round out the chemical barriers.

Pattern recognition. Infected and sentinel cells detect PAMPs (pathogen-associated molecular patterns) with PRRs (pattern recognition receptors). For influenza, four sensors matter — three that detect viral RNA and one that senses the damage the virus does:

Interferon is the central antiviral signal. Infected cells secrete type I interferons (IFN-α and IFN-β). These act mostly on neighboring cells, binding the IFNAR receptor and switching on, through the JAK–STAT pathway, several hundred interferon-stimulated genes (ISGs) that put the cell into an antiviral state before it is infected. The ones tests name:

Influenza fights back with NS1, which binds dsRNA so sensors cannot see it, blocks RIG-I signalling, and interferes with host mRNA processing and export. NS1 is the reason an otherwise healthy interferon system does not stop the infection outright, and deleting it is one way live attenuated vaccines are designed.

Cells and cascades. Alveolar macrophages and neutrophils phagocytose virions, debris and infected cells. Natural killer (NK) cells kill infected cells without needing MHC-restricted recognition: they read the balance of activating signals against inhibitory signals from MHC class I, so a cell that has lowered its MHC I to hide from T cells ("missing self") is killed by NK cells instead. NK cells also perform ADCC (antibody-dependent cellular cytotoxicity) by binding, through their CD16 Fc receptor, IgG that is already stuck to HA on the infected cell's surface — a bridge from the adaptive side back to the innate one.

Complement is a cascade of about 30 plasma proteins with three activation routes: the classical pathway (C1q binding antibody already attached to antigen), the lectin pathway (mannose-binding lectin recognizing the sugars on HA's glycan shield), and the alternative pathway (spontaneous C3 hydrolysis on any surface lacking host regulators). All three converge on C3 convertase, and the shared outputs are opsonization (C3b tagging particles for phagocytes), chemotaxis and inflammation (the anaphylatoxins C3a and C5a), and the membrane attack complex (C5b–C9), which punches pores in membranes. Note that complement can neutralize an enveloped virion directly, which naked viruses largely escape.

Inflammation and symptoms. Infected cells and macrophages release cytokines — IL-1, IL-6, TNF-α — and chemokines that recruit more cells. IL-1 and IL-6 act on the hypothalamus to raise the thermostat, producing fever; prostaglandin E₂ is the final mediator, which is why aspirin and ibuprofen (COX inhibitors) lower a fever. Fever, aches and fatigue largely come from these signals, not from the virus directly. When the response runs out of control — most often in severe influenza — massive cytokine release ("cytokine storm") damages the alveolar barrier and causes the acute respiratory distress that kills. This is the mechanistic answer to "why were healthy young adults at risk in 1918?": a vigorous immune system can itself do the damage.

Adaptive immunity

Adaptive immunity is slow the first time because it works by clonal selection: the body makes an enormous repertoire of lymphocytes, each with one randomly generated receptor, and an antigen selects the few that fit, which then divide. That takes about a week. The second time, the expanded memory clones are already there, so the response is faster, larger and of higher affinity.

Antigen presentation and MHC

T cells cannot see a whole virus. They see short peptides held in the groove of a major histocompatibility complex (MHC) molecule — in humans, HLA. There are two pathways, and the pairing with the T cell co-receptor is the fact most asked:

Mnemonic: multiply to eight. CD8 × MHC I = 8; CD4 × MHC II = 8.

A CD8⁺ T cell that finds its peptide kills by releasing perforin (which lets granzymes in) and granzymes (proteases that trigger apoptosis), and by engaging Fas on the target. Killing the infected cell before it finishes making virions is how cell-mediated immunity limits an established influenza infection. Because the peptides come from whatever protein is being made, many of them come from conserved internal proteins (NP, M1) rather than from the variable HA head — so T-cell immunity cross-protects between strains and even subtypes far better than antibody immunity does, which is why previous infection often reduces severity even when it does not prevent infection.

B cells, antibodies and affinity maturation

A B cell's receptor is a membrane-bound antibody. When it binds its antigen and gets help from a CD4⁺ T cell recognizing the same antigen on the B cell's MHC II, the B cell proliferates in a germinal center, where two things happen:

• Somatic hypermutation, carried out by the enzyme AID, scatters point mutations through the variable-region genes. B cells whose mutated receptors bind antigen better capture more of it, get more T-cell help, and survive; the rest die. Over successive rounds, average antibody affinity climbs — affinity maturation.
• Class switch recombination swaps the heavy-chain constant region (IgM → IgG, IgA or IgE) while leaving the variable region, and therefore the specificity, unchanged. Cytokines from the helper T cell choose the class.

The output is plasma cells (antibody factories, no longer displaying surface receptor) and memory B cells.

An antibody is a protein-structure question in disguise, so expect it on this event's tests. The basic unit is two identical heavy chains and two identical light chains, held together by interchain disulfide bonds in a Y shape. Each chain is a string of immunoglobulin folds — compact β-sandwich domains, two antiparallel sheets with Greek-key topology pinned face to face by an internal disulfide (a sandwich, not a barrel: the sheets stack, they do not close into a cylinder). Papain digestion cuts above the hinge into two Fab fragments (one binding site each) and one Fc; pepsin cuts below it, giving a bivalent F(ab′)₂. The actual contact surface, the paratope, is built from six hypervariable loops — the complementarity-determining regions (CDRs), three from VH and three from VL — and the patch of antigen it touches is the epitope. A conformational epitope, like most of those on HA, exists only in the folded protein, which is why denaturing the antigen can abolish antibody binding.

How antibodies actually stop influenza

"Neutralization" is not one mechanism. Against HA there are several, and a good test question asks which one a given antibody uses:

1. Block receptor binding. An antibody whose footprint covers or overlaps the receptor binding site physically prevents HA from reaching sialic acid. This is the mechanism HAI measures.
2. Block fusion. Stem antibodies let the virus bind and be endocytosed but clamp HA2 so it cannot do its spring-loaded refolding; some also prevent HA0 cleavage.
3. Aggregation. Multivalent antibodies, especially IgM, cross-link virions into clumps that cannot infect and are easily cleared.
4. Opsonization and complement. Fc regions on bound antibody recruit phagocytes (via Fcγ receptors) and C1q, destroying the particle or the infected cell.
5. ADCC. IgG bound to HA on an infected cell's surface is read by NK cell CD16, which triggers killing.

Most neutralizing antibodies bind the HA head, around and over the receptor binding site, because it is exposed and immunodominant. On H1 heads these targets are called antigenic sites Sa, Sb, Ca1, Ca2 and Cb (on H3 they are labeled A–E). The head tolerates mutation, so these sites drift. The conserved pocket floor cannot change much without losing sialic acid binding, but it is a small, shallow pocket that most antibodies cannot reach deeply; HA also hides variable surfaces behind N-linked glycans, and adding a new glycosylation site is one of the cheapest ways to escape an antibody. The stem is far more conserved, and broadly neutralizing stem antibodies — which often cross-react within a phylogenetic group (group 1 includes H1, H2 and H5; group 2 includes H3 and H7) — are a major target of "universal" vaccine research.

The hemagglutination inhibition (HAI or HI) assay measures anti-HA antibodies: HA on virions crosslinks red blood cells by binding their sialic acids (that is where the name hemagglutinin comes from), and serum antibodies that block the binding site prevent the clumping. Un-agglutinated red cells roll down a round-bottomed well into a sharp button; agglutinated ones form a diffuse mat.

Immune memory and original antigenic sin

Memory is why adaptive immunity is worth its slow start. A secondary response has a shorter lag, a higher peak, mostly IgG rather than IgM, and higher average affinity thanks to earlier affinity maturation.

Memory also has a downside peculiar to influenza. Original antigenic sin, now usually called immune imprinting, is the tendency of the immune system to answer a new, related strain by recalling memory cells raised against the first strain it ever met, rather than by generating fresh responses to the new epitopes. The recalled antibodies may fit the new HA poorly. Consequences the exam cares about:

• Your birth year predicts which HA group you respond to best, because it determines the subtype of your first influenza infection.
• In 2009, people born before about 1957 had 1918-descended H1 memory that cross-reacted with the pandemic HA, so severe disease skewed young — the opposite of a normal season.
• Imprinting also partly explains why repeated annual vaccination with similar strains can give diminishing boosts.

How influenza spreads through a population

Protein Modeling's exam asks about "the viral infection process", and the easiest population-level questions follow directly from the biology above.

Prevention and treatment

Influenza vaccines mostly work by raising anti-HA antibody, so the whole enterprise turns on predicting HA. Strain selection is done by WHO consultations twice a year — about February for the Northern Hemisphere season and September for the Southern — using global surveillance data, which leaves roughly six months for manufacture. Vaccines have been trivalent since the 2024–25 season, after B/Yamagata dropped out of circulation: an H1N1, an H3N2 and a B/Victoria component. A mismatch happens when the circulating virus drifts after the strains are chosen, and that, rather than any failure of the immune response, is the usual cause of a low-effectiveness season.

Inactivated and recombinant vaccines are given intramuscularly and raise mostly serum IgG, which is good at preventing severe disease; LAIV raises mucosal IgA at the site of infection, which is better at blocking transmission. Vaccination cannot cause influenza: inactivated and recombinant vaccines contain no replicating virus, and the soreness and low fever some people get are the innate response to the injected antigen.

Antivirals are a second line, most useful when started within about 48 hours of symptom onset.

Hemagglutinin and sialic acid receptors

This is the heart of the 2027 model and, at State and Nationals, a named exam topic.

Sequence to infection, in one chain of cause and effect

This is the synthesis a State or National question is really asking for, and the one thing to be able to recite while pointing at your model.

Sialic acid, the receptor

Sialic acids are nine-carbon acidic sugars that cap the ends of glycan chains on cell-surface glycoproteins and glycolipids. The most common in humans is N-acetylneuraminic acid (Neu5Ac). Its key chemical features are the ones HA reads:

• a carboxylate at C1, negatively charged at physiological pH (pKa about 2.6), which makes sialic acid an acidic sugar and gives cell surfaces part of their negative charge;
• an N-acetyl group at C5, whose methyl packs against a nonpolar surface;
• a glycerol side chain (C7–C9) with hydroxyls that hydrogen-bond;
• the C2 carbon, which links to the next sugar (usually galactose) through a glycosidic bond.

A detail worth knowing: many animals, including pigs and horses, also make Neu5Gc (N-glycolylneuraminic acid, with an extra oxygen); humans cannot, because the CMAH gene is inactivated in our lineage. Human airway sialic acid is therefore essentially all Neu5Ac, which is one more species-level filter on which viruses bind us well.

α2,3 versus α2,6 linkages

The bond from sialic acid's C2 to galactose can go to galactose carbon 3 (α2,3) or carbon 6 (α2,6). This one detail controls host range.

Pigs carry both linkages in their airways, which is why they can be infected by both bird and human strains and can act as a mixing vessel for reassortment, as happened in 2009.

The linkage is not the whole story. In the glycan topology model (Chandrasekaran et al., 2008), what matters is the three-dimensional shape the whole glycan adopts: short glycans of either linkage fold into a cone, whereas long α2,6 glycans can fold into an umbrella that drapes over a wider patch of the HA surface. The extra C5–C6 bond in an α2,6 linkage gives the junction more rotational freedom, so only α2,6 chains can make the umbrella. Human-adapted HAs bind the umbrella form of long α2,6 glycans; avian HAs, including H5, do not, which is offered as one reason avian viruses can bind some human α2,6 receptors in the lab yet still fail to transmit. The practical exam version: linkage sets the shape, and HA reads shape, not just the bond.

The receptor binding site

The binding site is a shallow pocket near the top of each HA1 head — three identical sites per trimer, one per protomer. Its floor is formed by a set of conserved residues, usually listed as Tyr98, Trp153, His183 and Tyr195, and its rim by four structural elements: the 130-loop, 150-loop, 190-helix and 220-loop.

In 3UBE the ligand is LSTc, a human-type α2,6-linked pentasaccharide, bound to the A/California/04/2009 HA at 2.15 Å resolution. The companion structures in the same study are 3UBN (6′-SLN, a short α2,6 trisaccharide), 3UBJ (LSTa, avian-type α2,3) and 3UBQ (3′-SLN). The construct carries an engineered G205C/R220C disulfide across the subunit interface, added purely to stabilize the trimer for crystallography — a detail to mention in a legend, because those two cysteines are not in the wild-type sequence. In the file, chain A is HA1 and chain B is HA2, and the segment you build is chain A residues 153–252.

The contacts measured in the 3UBE coordinates themselves, sialic acid first and then the galactose beyond it:

Read the pattern rather than memorizing every number. The carboxylate is gripped by the 130-loop (backbone N of 137, side chain of 136) and by Gln226; the glycerol hydroxyls hydrogen-bond to Tyr98 and His183 on the floor; the N-acetyl methyl sits on the flat face of Trp153; and only the residues of the 220-loop and 190-helix reach past sialic acid to the galactose and beyond. Note also how many of these are backbone contacts (Val135 O, Ala137 N, Asp225 O): backbone atoms cannot mutate away, which is part of why the core of the site is conserved while its rim drifts.

Water matters too. Several contacts in this site are water-mediated rather than direct, including bridges from Ser186 and Asp190 to the glycerol hydroxyl and from Lys133A to the 4-hydroxyl — the last of these reported as a feature unique to H1 HAs. A single ordered water molecule in a 2.15 Å structure is a real, modeled atom, which is why high resolution matters for this kind of analysis.

What sets receptor preference in H1 HAs

In human-adapted H1 viruses, including 1918 and 2009, two aspartates matter most: Asp190 (on the 190-helix) and Asp225 (in the 220-loop), in H3 numbering. Avian H1 viruses usually have glutamate at 190 and glycine at 225, so the human-adaptation pair is written E190D / G225D. Their job is to reach past the terminal sialic acid: Asp225 hydrogen-bonds the galactose, and Asp190 is the only residue that contacts the glycan beyond galactose at all, hydrogen-bonding the acetamido nitrogen of GlcNAc-3 and the 2-hydroxyl of Gal-4. An α2,6 glycan, folded compactly with a cis linkage between sialic acid and galactose, lays those sugars where the aspartates can reach them; an α2,3 glycan, with its different geometry, cannot satisfy them.

The crystallography shows the difference directly. With LSTc (α2,6) the sugar is well ordered in every binding site in the crystal and can be modeled out to the fourth sugar. With LSTa (α2,3) the density is weak in four of six sites and absent in the other two, and only the sialic acid and the galactose beyond it could be modeled. Disorder in a crystal is the structural signature of weak, non-specific binding.

A third residue completes the picture: Lys222 is held in a fixed orientation by a small charge network with Asp225 and Glu227, which positions its side chain to hydrogen-bond the galactose.

The two solutions are different chemistry for the same problem, and tests like to pair them: 190/225 for H1, 226/228 for H2 and H3. Both change the 190-helix/220-loop rim so that an α2,6 glycan fits and an α2,3 glycan does not.

D225G (H1 numbering D222G) is the variant to know from 2009. It removes the favourable hydrogen bond to galactose, which lowers α2,6 avidity, and makes the 220-loop more flexible, which improves α2,3 binding. Viruses carrying it were reported more often in severe and fatal cases, consistent with better binding deep in the lung where α2,3 receptors are, and it tends to reduce transmissibility. Its effect in the 2009 HA appears milder than the same change in the 1918 HA — possibly because Glu227 can partly compensate for the lost aspartate and keep Lys222 and the 220-loop in place. That contrast, the same substitution behaving differently in two related HAs, is exactly the kind of structure–function reasoning a National-level question wants: a residue's effect depends on its whole network, not on its identity alone.

Avidity and the HA–NA balance

Each HA–sialic acid contact is weak. A single HA binding one sialoside has a dissociation constant in the millimolar range — about a million times weaker than a typical nanomolar antibody–antigen interaction — and the Xu study likewise found 2009 HA binding α2,6 disaccharides more than tenfold more weakly than related strains in ELISA, with α2,3 binding barely detectable. The virus sticks anyway through avidity: a virion carries hundreds of HA trimers, each with three sites, and dozens of them engage sialic acids on the cell at once. Many weak contacts in parallel give an effective attachment that is many orders of magnitude tighter than any one of them, because all of them must let go simultaneously for the virion to drift away.

Avidity is also why NA is essential rather than a luxury. If HA binds too tightly or NA cuts too slowly, new virions are recaptured by sialic acid on the cell they just left, or trapped in sialylated mucus; if NA cuts too fast or HA binds too weakly, virions cannot hold on long enough to be taken in. Viruses therefore evolve a matched HA–NA balance, and changing one forces compensatory change in the other. Two testable consequences: drifting HA to escape antibodies often requires a compensating NA change, and a virus made resistant to oseltamivir (weakened NA) frequently acquires HA mutations that lower receptor avidity to restore the balance.

Structures, prediction and the PDB

The Protein Data Bank

The Protein Data Bank (PDB), founded in 1971 at Brookhaven with seven structures, is the single worldwide archive of experimentally determined 3D structures of proteins and nucleic acids. It now holds more than 250,000 experimental entries and is run by the wwPDB partners (RCSB PDB in the US, PDBe in Europe, PDBj in Japan, plus the NMR and EM data banks). Each entry has a four-character ID such as 3UBE, its coordinates, the experimental data and metadata. The archive's standard format is now PDBx/mmCIF, because the older fixed-column .pdb format cannot hold more than 99,999 atoms; viewers such as JUDE read both.

A PDB file is a text file. The records you should recognize:

The table's two central rows, ATOM and HETATM, are what a viewer draws; everything else is context. Note that CA in the atom-name column means the alpha carbon, not calcium (calcium appears as a HETATM ion named CA with element Ca), a classic trap when selecting atoms.

X-ray crystallography

How it works.

1. Express and purify milligrams of pure, homogeneous protein.
2. Crystallize it. Proteins are slowly brought out of solution (by adding a precipitant such as polyethylene glycol or ammonium sulfate, often by vapor diffusion in a hanging or sitting drop) until molecules pack into an ordered three-dimensional lattice. The first step is nucleation, the formation of a tiny ordered seed, followed by growth. Crystals form because molecules can lower their free energy by making regular, repeated contacts (hydrogen bonds, salt bridges, van der Waals contacts) with neighbors; they are held only by these weak contacts and contain roughly 30–70% solvent, which is why protein crystals are fragile and the proteins inside remain in a near-native state. Finding conditions is trial and error and is often the slowest step; flexible or membrane proteins may never crystallize.
3. Collect diffraction: a beam of X-rays (wavelength about 1 Å, similar to bond lengths) hits the crystal, usually at a synchrotron and with the crystal frozen to limit radiation damage. Electrons scatter the X-rays, and the regular lattice makes the scattered waves reinforce only in particular directions, giving a pattern of spots (reflections). Their positions give the lattice geometry; their intensities depend on the arrangement of electrons in the repeating unit.
4. Solve the phase problem. Detectors record each spot's intensity (amplitude) but not its phase, and both are needed to calculate the image. Phases come from molecular replacement (using a similar known structure as a starting model), heavy-atom derivatives (isomorphous replacement), or anomalous scattering (SAD/MAD, often with selenomethionine in place of Met).
5. Calculate an electron density map (a Fourier transform of amplitudes plus phases), build the model into it using the known sequence, and refine it against the data.

Diffraction obeys Bragg's law, the principle used to turn spot positions into lattice spacings.

Quality measures, and what each tells you:

The table moves from global quality (resolution, R, R-free) to atom-by-atom reliability (B-factor, occupancy). In a viewer, coloring by temperature (B-factor) shows rigid regions in blue and mobile regions in red; loops and the protein surface are usually hottest.

X-ray strengths and weaknesses. Strengths: atomic or near-atomic detail, no size limit in principle, high throughput once crystals exist, and clear views of bound ligands and water (most drug-design structures are crystallographic). Weaknesses: you must grow crystals, which can take months or fail; the result is a time-averaged static picture; crystal packing can distort surface loops; flexible regions and glycans may be missing; hydrogens are usually invisible; and the phase problem must be solved.

NMR spectroscopy

Nuclear magnetic resonance places a protein solution in a strong magnetic field and measures how nuclei with spin (¹H, and ¹⁵N and ¹³C introduced by isotope labeling) absorb radio waves. Each nucleus's chemical shift depends on its local environment, and NOE signals reveal pairs of hydrogens closer than about 5–6 Å. These distance and angle restraints are used to calculate a family of structures that all satisfy the data, so an NMR PDB entry contains an ensemble of models (often 20). Where the models agree the protein is rigid; where they spread out it is flexible.

• Advantages: no crystal needed; protein in solution near physiological conditions; directly measures dynamics and flexibility; can follow folding, binding and pH titrations.
• Limitations: practical mostly for small proteins (roughly under 30–50 kDa, larger with special labeling); needs concentrated, stable, isotope-labeled samples; generally lower precision than high-resolution X-ray; slow data analysis.

Two NMR effects show up in questions. In hydrogen–deuterium exchange, a protein dissolved in D₂O gradually loses the ¹H signals of its exchangeable amide protons as they are replaced by deuterium (which is not seen in ¹H NMR); amides on the surface exchange fast, while those buried in the core or locked in hydrogen bonds of helices and sheets exchange slowly, so the exchange rate maps structure and stability. And aromatic rings create ring-current shifts on nearby protons in a folded protein; those unusual shifts disappear when the protein unfolds and the environments average out.

Cryo-electron microscopy

In cryo-EM, a thin film of protein solution on a grid is plunged into liquid ethane so fast that water freezes as glass-like vitreous ice, trapping thousands of molecules in random orientations. A transmission electron microscope records many low-dose images (movies, called micrographs). Software then picks particles, sorts and averages particles with the same orientation into 2D class averages, combines the views into a 3D reconstruction (density map), and an atomic model is built into the map.

• Advantages: no crystals; tiny amounts of sample; ideal for large complexes, membrane proteins and flexible machines; can separate several conformations present in one sample. Resolution is now routinely 2–4 Å and better than 1.5 Å for the best samples.
• Limitations: small proteins (below roughly 50–100 kDa) give too little contrast; expensive microscopes; resolution often varies across one map (flexible edges are blurrier).

The 2017 Nobel Prize in Chemistry (Jacques Dubochet, Joachim Frank, Richard Henderson) recognized cryo-EM. Whole HA trimers and HA–antibody complexes are now routinely solved by cryo-EM, while HA ectodomains and heads with bound receptor analogs, like 3UBE (a full HA1/HA2 ectodomain), are mostly crystal structures.

Comparing the three methods

The table summarizes the trade-offs; questions usually give a protein's size or properties and ask which method fits. A 155 kDa soluble enzyme that crystallizes: X-ray (or cryo-EM). A 12 kDa flexible domain: NMR. A 1 MDa membrane complex: cryo-EM.

Other techniques measure structure without giving atomic coordinates: circular dichroism (CD) in the far-UV (about 190–250 nm) estimates secondary structure (α-helix shows minima at 208 and 222 nm; β-sheet a minimum near 218 nm) and follows unfolding; fluorescence of tryptophan reports whether Trp is buried or exposed; mass spectrometry gives mass, sequence and modifications; small-angle X-ray scattering (SAXS) gives overall shape in solution. Edman degradation sequences a protein from the N-terminus one residue at a time; historically, crystal structures were interpreted by combining diffraction with the separately determined amino-acid sequence.

A short history

The table is roughly the sequence tests draw from; the HA row ties the history to your protein.

Predicting and modeling structures

Anfinsen showed that a protein's sequence contains the information for its fold, so in principle structure can be computed from sequence. Methods, from most to least dependent on known structures:

• Homology (comparative) modeling: if the target's sequence is similar to a protein of known structure (a template, generally above about 30% identity), align the two sequences and build the target on the template's backbone, then model loops and side chains. Works because structure is conserved more than sequence. SWISS-MODEL is a common server.
• Threading (fold recognition): test the sequence against a library of known folds to find one it fits, even without obvious sequence similarity.
• Ab initio (de novo) prediction and molecular dynamics: start from the sequence or an extended chain and search for the lowest-energy structure using physics-based energy functions. Molecular dynamics (MD) simulates atomic motion over time using a force field (CHARMM, AMBER) whose terms include bond stretching, angle bending, torsion (dihedral) rotation, van der Waals interactions (Lennard-Jones potential) and electrostatics (Coulomb's law). MD is good for dynamics but too slow to fold most proteins from scratch.
• Deep learning (AlphaFold, RoseTTAFold): networks trained on the PDB and on multiple sequence alignments, which reveal pairs of positions that mutate together (co-evolution, a hint that they touch).

Most pipelines share four stages: backbone prediction, loop modeling, side-chain packing, and refinement.

Molecular replacement uses the same idea in crystallography: a known structure of a similar protein is placed in the new crystal's unit cell to supply initial phases, and the model is then rebuilt and refined against the new data. 3UBE was solved this way from another 2009 HA structure.

AlphaFold (Google DeepMind; John Jumper, Demis Hassabis and colleagues) predicts a 3D structure from an amino-acid sequence alone. AlphaFold2 achieved near-experimental accuracy for many proteins at CASP14 in 2020, and the AlphaFold Protein Structure Database (with EMBL-EBI) now holds over 200 million predicted structures, nearly every known protein sequence, freely available. AlphaFold 3 (2024) extends prediction to complexes with nucleic acids, ligands and ions.

pLDDT (predicted local distance difference test) reports local confidence for each residue; a separate PAE (predicted aligned error) plot reports confidence in the relative placement of domains. Limits: predictions are a single static model; they may be wrong for disordered regions, alternative conformations, the effect of a point mutation, or a protein's complex with a ligand; and they are not experimental evidence.

RMSD (root-mean-square deviation) is the standard single number for how similar two structures are. After the structures are optimally superimposed,

$$\text{RMSD} = \sqrt{\frac{1}{N}\sum_{i=1}^{N} d_i^{\,2}}$$

where $d_i$ is the distance between the i-th pair of matching atoms (usually Cα). Below about 1 Å the structures are nearly identical; 1–2 Å means the same fold with local differences; values above about 3–4 Å mean substantial differences. Because the differences are squared, a few large deviations dominate.

Distance matrices and contact maps represent a structure by the distances (or contacts below a cutoff such as 8 Å) between every pair of residues. They depend only on relative positions, so they are invariant to rotation and translation of the whole molecule, which makes them convenient inputs for prediction networks. The matrix is symmetric (entry i,j equals entry j,i) with zeros on the main diagonal; helices appear as bands next to the diagonal and sheets as lines perpendicular (antiparallel) or parallel to it. Distances are also unchanged by mirror reflection, so a distance matrix alone cannot tell a right-handed helix from a left-handed one.

De novo designed proteins ("designer proteins") are new sequences computed to fold into chosen shapes, pioneered by David Baker's group (Top7, 2003, a fold not seen in nature). Today they include designed binders that block HA's stem or receptor site, vaccine scaffolds and enzymes.

How structure-method questions are asked

• Advantage and disadvantage of X-ray, NMR or cryo-EM, or which method fits a given protein size.
• Steps of crystallography (crystallize → diffract → phase → map → model) or cryo-EM (freeze on grid → image → pick particles → 2D classes → 3D map → model).
• Definitions: resolution, R-factor, B-factor, Matthews coefficient, pLDDT, RMSD, molecular replacement, homology modeling.
• Bragg's law calculations.
• History: Sumner, Kendrew and Perutz, PDB founding, Nobel prizes (1962, 2017, 2024).
• PDB file contents: ATOM vs HETATM, HELIX and SHEET records, chain IDs.

Working in JUDE

JUDE is the browser version of Jmol that the 2027 rules name for both the build and the exam. It is free at the 3D Molecular Designs digital modeling hub, and it is the only website you may use during the event, so practice until the common operations are muscle memory. Under the menus, JUDE runs Jmol/JSmol, the open-source molecular viewer that reads PDB and mmCIF files, so anything you learn about Jmol commands carries over and questions are often written in Jmol syntax.

Molecular visualization matters for this event because a sequence or a flat picture hides the spatial relationships the exam asks about: which residues touch, which face the core, how a ligand sits in a pocket. JUDE lets you identify chains, residues, secondary structures, side chains and ligands and measure the distances between them.

The table lists the operations in the order you usually perform them. The command reference below gives the Jmol syntax behind each.

Representations

Each representation shows some things and hides others; tests ask for advantages and disadvantages.

The rows run roughly from least to most atomic detail. A common combination for binding questions is cartoon for the protein plus ball-and-stick for the pocket side chains and the ligand.

Jmol command reference

Jmol commands work in two steps: select a set of atoms, then apply display, color, label or measurement commands, which act only on the current selection. Commands end at a line break or a semicolon. Selecting does not change the picture by itself.

ReferenceJmol commands for Protein Modeling· 25 rows

:A means chain A; 153-252 is a residue-number range; .CA names an atom. Commands are not case-sensitive.

Read the table by row when building a command and by the Notes column when a question asks why a command misbehaves.

25 rows

Purpose	Command	Notes
Select everything	select all (or select *)	Resets the selection before a new display command
Select a residue in one chain	select 190:A	Without :A, every chain's residue 190 is selected
Select a range in one chain	select 153-252:A	Your model's segment
Select a whole chain	select :A	Chain A is HA1 in 3UBE
Select by residue type	select TRP, select cys	Three-letter names
Select one atom	select 153:A.NE1	Atom names as in the PDB file; CA = α-carbon
Predefined sets	select protein, ligand, hetero, water, solvent, helix, sheet, backbone, sidechain, acidic, basic, charged, polar, hydrophobic, aromatic, cystine	Combine with and, or, not. Jmol's hydrophobic set is Ala, Gly, Ile, Leu, Met, Phe, Pro, Trp, Tyr, Val; polar is every other amino acid
Combine conditions	select polar and sidechain, select hetero and not water	and, or, not; parentheses group them
Neighbors of a ligand	select within(5.0, ligand)	Everything within 5 Å (the distance in within() is always in Å)
Show only the selection	restrict 153-252:A	Hides everything else (selects and hides)
Show or hide	display selected, hide water	display and hide act without discarding the selection
Representations	trace, backbone 0.4, cartoon, ribbon, rockets, wireframe 0.15, spacefill, spacefill 50%, spacefill off	Apply to the current selection
Color by element	color cpk	C gray, O red, N blue, S yellow
Color by secondary structure	color structure	Helix magenta-pink, strand yellow, turn blue, other white
Color by chain	color chain	Each chain a different color
Color by residue type	color amino	Asp/Glu red, Lys/Arg blue, Cys/Met yellow, Ser/Thr orange, Phe/Tyr dark blue, Asn/Gln cyan, Leu/Ile/Val green, Gly white, Ala gray, Trp purple-pink, His pale blue, Pro salmon
Rainbow N→C	color group	Blue at the N-terminus to red at the C-terminus
Color by B-factor	color temperature	Blue = low B (rigid), red = high B (mobile)
Size by B-factor	spacefill temperature	Bigger spheres for more mobile atoms
Any color	color yellow, color [255,128,0]	Named colors or an RGB triple
Label	label %n%R	Residue name and number; label off
Measure distance	measure (153:A.NE1) (190:A.OD1)	Or double-click two atoms; result in Å
Hydrogen bonds	calculate hbonds then hbonds on	Backbone H-bonds are calculated from structure; ligand H-bonds may need measuring
Disulfides	ssbonds on	Draws S–S bonds; ssbonds 0.2 sets their thickness
View	center selected, zoom 200, spin on, reset	reset returns to the default orientation

The two lines that cause the most trouble on tests:

• Integer versus decimal. Jmol inherited RasMol units from older software: an integer width or radius is in units of 1/250 Å, while a number with a decimal point is in ångströms. backbone 1 draws a thread 0.004 Å thick (almost invisible); backbone 1.0 draws a 1 Å rod. Likewise backbone 100 equals backbone 0.4. The same rule applies to wireframe, spacefill and trace widths; distances inside within() are always in Å.
• Chain ambiguity. A structure with several chains (3UBE has several copies of HA) has a residue 153 in each chain. select 153 grabs all of them; select 153:A grabs only chain A's.

Colors to know

Element coloring follows the CPK convention (named for Corey, Pauling and Koltun's physical models): carbon gray or black, oxygen red, nitrogen blue, hydrogen white, sulfur yellow, phosphorus orange, iron orange-brown, chlorine green, sodium violet, calcium green, zinc slate blue-gray. Representation styles are often confused, so learn what each one hides: a cartoon invents ribbons that are not atoms at all, and a trace draws only α-carbons, exactly like your physical model.

Model colors (your legend) and Jmol's default colors are different systems: your model might show hydrophobic side chains yellow while Jmol's color amino shows them green. Say which scheme you mean.

Ligands, water and other molecules

Ligands, sugars, ions and water are stored as HETATM records and are hidden or shown differently from the protein by default. select ligand (or select hetero and not water) finds them; select within(4.0, ligand) and sidechain lists the side chains that contact them. In 3UBE, look for N-acetylglucosamine (NAG) of the N-linked glycans attached to Asn residues (the LSTc chains also contain a NAG unit, so check whether a NAG is bonded to Asn), and the LSTc receptor analog (sialic acid, galactose and N-acetylglucosamine units) sitting in the receptor-binding pocket. Water appears as isolated red oxygen spheres (hydrogens are not in X-ray files).

Jmol computes backbone hydrogen bonds for secondary structure but does not automatically draw every protein–ligand hydrogen bond, so measure: if a donor N or O and an acceptor O or N are within about 3.5 Å (some tests use 3.4 Å), count it as a hydrogen bond; typical values are 2.7–3.2 Å. Other useful reference distances for measurements:

These distances let you name an interaction from a measurement. Anything farther than about 5 Å is not a direct contact, though it may still be mediated by water.

Reading the chemical environment of a side chain

A common exam task gives you a residue and asks why it sits where it does. Work from the atoms outward:

1. What is the side chain made of? Carbons only means hydrophobic; N or O means hydrogen bonding; charged groups mean salt bridges.
2. What is within a few ångströms? Other side chains, backbone atoms, water, or a ligand.
3. Name the interaction and say what it stabilizes: a buried hydrophobic cluster, a surface salt bridge, a disulfide bond tying two loops together, or a contact that holds a ligand.
4. Predict a mutation's effect: putting a charged residue into a buried hydrophobic cluster destabilizes the fold; removing a cysteine breaks a disulfide; changing a pocket residue changes binding, not folding.

In JUDE, steps 1–3 translate into a short command sequence: select the residue and everything within about 4–5 Å (select within(4.5, 190:A)), show it as ball-and-stick colored by element, label the residues, and measure the closest N/O pairs. Spacefill of the same selection tells you whether the side chain is buried (no water can reach it) or exposed.

How visualization questions are asked

• Identify helices and strands in a residue range and list their first and last residues (use color structure or the HELIX/SHEET records).
• Name the residue(s) contacting a ligand, and the interaction type, from distances you measure.
• Write a command sequence that selects, displays and colors a feature.
• Compare representations: advantages and disadvantages of cartoon, wireframe, spacefill, trace.
• Explain a quirk: integer vs decimal units, chain ambiguity, why hydrogens are absent, what temperature coloring shows.

Strategy

• Start the model early. It must be finished before your first tournament and it travels to every level. Build it, then improve its features between tournaments as the rubric gets stricter.
• Split the preparation, not the knowledge. One partner can own the build and the JUDE operations while the other owns influenza biology and immunity, but both must be able to answer either half: the exam is 60% of the score and you sit it together.
• Put structure-specific facts on your sheets. The 3UBE segment's helix and strand boundaries, your legend's color key, the binding-site residues and loops, the α2,3 versus α2,6 contrast, the infection cycle in order, the innate and adaptive players, and the scale conversion (0.53 cm per Å).
• Rehearse JUDE under constraints. No search engine, one device, a borrowed network. Practice loading, selecting a residue range, measuring a distance and displaying a ligand without any notes.
• Protect the model in transit. It is impounded intact; a backbone that sags or a feature that falls off costs accuracy points no exam answer can win back. Build a stand and a box.
• Budget the 50 minutes. Leave a fixed block for the molecular exploration questions; they are the ones that need the device and are easiest to run out of time on.

Common pitfalls

One checklist of the mistakes that cost the most points, gathered from the pitfalls flagged through the guide plus the misreadings that show up most often in tournament answer keys. Read it the night before; each item is a complete answer on its own.

How tournaments test this

Three things matter about the question bank behind this guide. First, the archive's 1,962 Protein Modeling questions are dominated by the 2019-20 and 2020-21 seasons, when the event's protein was a CRISPR-associated nuclease. When we labeled the practice bank against the 2027 rules, only about three in five questions were still in scope; most of the rest are CRISPR, base editing, other retired target proteins or DNA-only molecular biology, and the October 2026 bank still tags 341 of its 1,874 questions CRISPR & Gene Editing. About 1,150 of those bank questions are exported from or match archive tests, and most of the rest come from older tournament tests the archive does not have. Practice on old tests, but skip those questions. Second, the question types are what the 2027 rules describe: about 54% free response, 32% multiple choice, 6% select-all, 4% true/false and 4% numeric, so expect to write sentences and expect select-all items with more than one correct option. Third, virus and immunity questions are almost absent from the old bank because the topic is new for 2027, while the rules give it half the exam's topic list: that is the gap between what you can practice and what you will be asked, and the influenza and immunity chapters here exist to fill it.

The table below shows roughly where the in-scope questions fell in that labeling, which is the best guide to where study time pays.

Shares are rounded to the nearest percent, so they add to a little over 100.

The last line is the one to watch. Virus and immunity questions are almost absent from the old bank because the topic is new for 2027, yet the rules list it as a major exam area and give it the whole second half of the written test's topic list. Expect the 2027 crop of tests to ask far more about influenza and immunity than any archived test did, and expect hemagglutinin, sialic acid, receptor binding and influenza to join the archive's most frequent words (protein, amino, structure, acid, residues).

Most common correct answers. Keys most often name tertiary (16 questions), hydrogen bonds (14), glycine (13), proline (13), isoleucine and lysine (12 each), cysteine (11), tyrosine (11), arginine (10), quaternary and secondary (9 each). Structure-level names and residues with distinctive side chains dominate, which is a good hint about what to over-learn: the special residues (Gly, Pro, Cys, His), the four structure levels, and the bond types.

What a typical 2027 question looks like. Expect: identify a structure level or a secondary-structure element from a description or a JUDE view; pick the residue with a given property; explain how a named interaction stabilizes a fold; compute a net charge or a pI; place a step of the infection cycle in order; explain what innate or adaptive immunity does to the virus; and, at State and Nationals, explain how HA's grip on sialic acid starts an infection and what changes receptor preference.

Test yourself

Real questions from public tournament tests in our archive, limited to the parts of the bank that stay inside the 2027 scope. Click an answer to check it; each links to the full test.

Test yourselfAmino acids and side chains

0 / 0 correct

1. 1

   Which of the following amino acids below is considered achiral?

   Question 1 options

   AMethionineBCysteineCTyrosineDGlycine

   From: University of Texas at Austin Regional, 2019-20, Division C

Practice more Protein Modeling questions →

Test yourselfCharge, pKa and ionization

0 / 0 correct

1. 1

   The C terminus of a peptide chain has what charge at physiological pH?

   Question 1 options

   A-2B-1C0D+1E+2

   From: University of Michigan Invitational, 2019-20, Division C

Practice more Protein Modeling questions →

Test yourselfPeptide bond and secondary structure

0 / 0 correct

1. 1

   Resonance causes peptide bonds in proteins to become

   Question 1 options

   AMore flexibleBNon-planarCLess reactive

   From: GGSO Invitational, 2020-21, Division C

Practice more Protein Modeling questions →

Test yourselfFolding, stability and higher-order structure

0 / 0 correct

1. 1

   Native proteins are generally far more stable than unfolded proteins, with the energy of these states often differing over a thousandfold.

   Question 1 options

   ATrueBFalse

   From: University of Michigan Invitational - MI, 2025-26, Division C

1 question from this set is no longer available.

Practice more Protein Modeling questions →

Test yourselfMutations, enzymes and viruses

0 / 0 correct

1. 1

   A researcher proposes a model of an enzyme-catalyzed reaction in which a reactant is converted to a product. The model is based on the idea that the reactant passes through a transition state within the enzyme-substrate complex before the reactant is converted to the product. Which of the following statements best helps explain how the enzyme speeds up the reaction?

   Question 1 options

   AThe enzyme’s active site binds to and stabilizes the reactant, which decreases the free-energy change of the reaction.BThe enzyme’s active site binds to and stabilizes the transition state, which decreases the activation energy of the reactionCThe enzyme’s active site binds to and stabilizes the product, which increases the amount of energy released by the reaction.DThe enzyme’s active site binds to and stabilizes both the reactant and the product at the same time, which increases the reaction’s equilibrium constant.

   From: Pearl City MiniSO Invitational, 2020-21, Division C

Practice more Protein Modeling questions →

Test yourselfStructure determination and visualization

0 / 0 correct

1. 1

   Scientists analyze protein folding in both a normal water environment and a deuterated heavy water environment. They find that the mass of the native state in both of these environments is different, with the mass in the deuterated water being slightly higher. Which of the following could be a plausible explanation?

   Question 1 options

   AHeavy water is associating to the protein in a way normal water does not due to significantly different reactivitiesBThe protein cannot fold properly in heavy water and is thus clustering a nonnegligible amount of heavy water in a hydrophilic pocket, leading to larger massCProteins interchange protons with their surrounding environments dynamically, and in the heavy water protons are exchanged for deuterium, leading to higher massDThe mass spectrometer is picking up signals from the deuterated water which could be hydrogen bonded to polar residues at the surface

   From: Golden Gate Invitational, 2019-20, Division C

Practice more Protein Modeling questions →

Breadth reference

Tournament tests ask a long tail of one-line facts: a term you met once, a bond length, a pKₐ, the size of an antibody. This chapter collects those facts in two large lookup tables, a glossary and a table of numbers, so you can check a word or a value in seconds instead of hunting through the guide. Both tables are filterable: type in the filter box above a table (for example "helix", "Jmol" or "kDa") and only matching rows stay visible. When you build a notesheet, filter by topic and copy only the rows you keep forgetting; the earlier chapters explain the ideas, and these tables are the compact restatement.

Glossary of protein science

The glossary runs roughly in the order of the guide: amino acids and bonds, structure from secondary to quaternary, folding and stability, function and binding, protein classes and modifications, mutation, structure methods, visualization, and finally the virology and immunology vocabulary an influenza-themed exam uses. The third column says where the term earns its points: the kind of question it answers, or the link to hemagglutinin (HA) and the 3UBE model.

ReferenceGlossary of protein science· 474 rows

Terms are grouped by topic, not alphabetized; use the filter box. HA = hemagglutinin; 3UBE = the 2027 model structure (2009 H1N1 HA with receptor analog LSTc).

Filter for a single word first; many terms appear in several rows (for example "disulfide" finds the bond, cystine, reducing agents, PDI, keratin and HA0).

474 rows

Term	What it means	Where it matters
Amino acid	Molecule with an amino group and a carboxylic acid group; in proteins both sit on the same (α) carbon with an H and a side chain	Name and functional-group questions ("why are they called amino acids")
α-carbon (Cα)	The central carbon bonded to the amino N, the carboxyl C, an H and the side chain	The one point per residue your model shows; C2 in IUPAC numbering
IUPAC numbering	Carboxyl carbon is C1 and the α-carbon C2, so α-amino acids are 2-aminocarboxylic acids; side-chain carbons are β, γ, δ, ε	Naming questions; "ε-amino group of Lys"
Side chain (R group)	The variable group on Cα that gives each amino acid its identity and chemistry	Classifying residues; the part of a residue that differs
Residue	An amino acid unit inside a chain, i.e. the free amino acid minus the water lost when the peptide bond formed	Mass calculations (residue mass = free mass − 18.02 Da)
Proteinogenic amino acid	One of the amino acids built into proteins by ribosomes: the 20 standard ones plus selenocysteine and pyrrolysine	"How many amino acids" questions (20 standard, 22 genetically encoded)
Selenocysteine (Sec, U)	Cysteine with Se in place of S; inserted at a UGA codon read through by a special tRNA and an mRNA signal	"21st amino acid"; nonstandard amino acid lists
Pyrrolysine (Pyl, O)	Lysine derivative carrying a pyrroline ring, encoded by UAG in some archaea and bacteria	"22nd amino acid"; the only side chain with a pyrroline ring
Essential amino acid	One the body cannot make in sufficient amounts and must eat: His, Ile, Leu, Lys, Met, Phe, Thr, Trp, Val	"Name the essential amino acids"; mnemonic PVT TIM HaLL
Conditionally essential amino acid	Normally made by the body but needed in the diet in illness, growth or prematurity (e.g., Arg, Cys, Gln, Gly, Pro, Tyr)	Distractor in essential/nonessential questions
Zwitterion	A molecule with both + and − charges and net charge zero; a free amino acid near neutral pH (–NH₃⁺ and –COO⁻)	Definition questions; why amino acids are water-soluble crystals with high melting points
Amphoteric	Able to act as both an acid and a base	Amino acids and proteins; buffering
Chiral center	An atom (usually C) with four different substituents, giving non-superimposable mirror images	All standard amino acids except glycine are chiral
Enantiomers	Non-superimposable mirror-image stereoisomers (L and D forms)	Why ribosomes need L-amino acids; mirror-world questions
Diastereomers	Stereoisomers that are not mirror images; arise with two chiral centers	Ile and Thr have two chiral centers, so 2² = 4 stereoisomers each
L/D configuration	Fischer-projection naming relative to glyceraldehyde; protein amino acids are L (amino group on the left)	"Most protein amino acids are L"; L-amino acids give right-handed α helices
R/S configuration	Cahn–Ingold–Prelog naming by substituent priority; most L-amino acids are S	L-cysteine is R because sulfur raises the side chain's priority
CORN rule	Looking down the H–Cα bond, CO–R–N read clockwise identifies an L-amino acid	Quick chirality check from a 3D drawing
Optical activity	Rotation of plane-polarized light by a chiral compound	Glycine, being achiral, does not rotate light
Aliphatic	Nonaromatic hydrocarbon side chain (Ala, Val, Leu, Ile; Pro; Met often grouped here)	Hydrophobic-core residues; Met as an "honorary aliphatic"
Aromatic	Side chain with a planar, conjugated ring: Phe, Tyr, Trp (His is an aromatic heterocycle)	UV absorbance at 280 nm, π-stacking, cation–π
β-branched	Side chain branched at Cβ (Val, Ile, Thr)	Favor β strands; restrict helix formation
Thiol (sulfhydryl)	–SH group (cysteine)	Forms disulfides by oxidation; binds Zn²⁺ and Fe
Thioether	C–S–C group (methionine)	Met's sulfur cannot form disulfides
Imidazole	Five-membered aromatic ring with two N atoms (histidine)	Side-chain pKₐ ≈ 6, so His titrates near physiological pH; binds Ni²⁺ in His-tags
Guanidinium	Planar, resonance-stabilized group of arginine, pKₐ ≈ 12.5	Arg is the most basic residue; binds phosphates and carboxylates
Indole	Benzene fused to a pyrrole (tryptophan)	Largest side chain; strongest 280 nm absorber; intrinsic fluorescence
Phenol	Aromatic ring bearing –OH (tyrosine)	Tyr phosphorylation; pKₐ ≈ 10
Carboxamide (amide side chain)	–CONH₂ of Asn and Gln; polar, not ionizable	Asn is the N-glycosylation site; amides are not basic
Imino acid (secondary amine)	Proline's ring bonds back to the backbone N, so Pro has no backbone N–H	Helix breaker; cannot donate a backbone H-bond
Polar uncharged	Side chain with O or N that H-bonds but carries no net charge at pH 7 (Ser, Thr, Asn, Gln, Cys, Tyr)	Surface residues; classification questions
Acidic residue	Asp and Glu: side-chain carboxylate, negatively charged at pH 7	Salt bridges; red in common color schemes
Basic residue	Lys, Arg, His: side chains that accept a proton; Lys and Arg positive at pH 7	DNA and sialic-acid-carboxylate binding; blue in color schemes
Hydropathy index	A numeric scale of side-chain hydrophobicity; Kyte–Doolittle runs from Ile +4.5 to Arg −4.5	Predicting buried residues and transmembrane helices
Amphipathic	Having a hydrophobic part and a hydrophilic part (a helix with a nonpolar face, a lipid)	Helical wheels; membrane-surface helices
Peptide bond	The amide C–N bond joining the carbonyl C of one residue to the amide N of the next	Primary structure; rigid, planar, almost always trans
Condensation (dehydration synthesis)	Reaction that joins two molecules with loss of water; forms each peptide bond	n residues are joined by n − 1 peptide bonds, releasing n − 1 waters
Hydrolysis	Splitting a bond by adding water; the reverse of condensation	Proteases; acid hydrolysis for composition analysis
Peptide, oligopeptide, polypeptide	Short chains (a few residues), up to about 20, and longer chains; "protein" usually means a folded, functional polypeptide (above about 50 residues)	Vocabulary questions
N-terminus	The chain end with the free α-amino group; sequences are written from it	Written first; +1 charge at pH 7; Edman degradation starts here
C-terminus	The chain end with the free α-carboxyl group	Written last; −1 charge at pH 7
Backbone (main chain)	The repeating N–Cα–C(=O) atoms of every residue	Secondary-structure H-bonds; trace and backbone displays
Peptide plane	The six atoms Cα–C(=O)–N(H)–Cα that lie in one plane because of resonance	Only φ and ψ rotate, so the backbone is a chain of rigid plates
Resonance (peptide)	Delocalization of the N lone pair into the C=O, giving the C–N bond about 40% double-bond character	Explains planarity, shorter C–N bond, low reactivity
Cis / trans peptide	The two isomers about the peptide bond (ω ≈ 0° or 180°); trans is favored about 1000 : 1 except before Pro	Cis-Pro in turns; peptidyl-prolyl isomerases
Isopeptide bond	An amide bond involving a side-chain amino or carboxyl group (e.g., Lys ε-NH₂ to ubiquitin's C-terminus)	Ubiquitination; cross-linked proteins
pKₐ	The pH at which a group is half protonated; $\mathrm{p}K_a = -\log K_a$	Charge-at-pH questions; seven residues have an ionizable side chain
Henderson–Hasselbalch equation	$\mathrm{pH} = \mathrm{p}K_a + \log([\mathrm{A^-}]/[\mathrm{HA}])$	Fraction protonated; one pH unit above pKₐ ≈ 91% deprotonated
Isoelectric point (pI)	pH at which a molecule's average net charge is zero; the average of the two pKₐ values flanking the neutral form	Isoelectric focusing; solubility is lowest at the pI
Titration curve	Plot of pH against added base; flat buffering regions center on each pKₐ	Reading pKₐ and pI off a graph
Strecker synthesis	Laboratory synthesis of α-amino acids from an aldehyde, ammonia and cyanide via an α-aminonitrile, then hydrolysis	Gives a racemic (D + L) mixture; chemistry-themed questions
Covalent bond	Bond formed by shared electron pairs; peptide and disulfide bonds in proteins	Strongest bonds; primary structure and cross-links
Noncovalent interaction	Weak interaction without shared electron pairs: H-bonds, ionic, van der Waals, hydrophobic	Hold secondary, tertiary and quaternary structure
Hydrogen bond	Attraction between an H on an electronegative donor (N, O) and a lone pair on an acceptor (N, O)	Secondary structure; specificity of binding
H-bond donor / acceptor	Donor = N–H or O–H group supplying the H; acceptor = O or N lone pair receiving it	Backbone N–H donates, C=O accepts; Pro cannot donate
Electronegativity	An atom's pull on shared electrons (F > O > N > C ≈ S > H)	Why C=O oxygen is δ− and amide H is δ+
Partial charge (δ)	Fractional charge on an atom from uneven electron sharing	Amide N–H and C=O dipoles; helix dipole
Salt bridge	Ionic interaction between oppositely charged side chains, often also H-bonded (Asp/Glu with Lys/Arg/His)	Tertiary and quaternary stability; mutations that remove a charge
Ionic (electrostatic) interaction	Coulomb attraction or repulsion between charges; weakened by high-dielectric water and salt	Charge-reversal mutations; electrostatic steering
Van der Waals interaction	Short-range attraction between all atoms at contact distance, with sharp repulsion if they overlap	Close packing of the core; contributes heavily in total
London dispersion force	The van der Waals attraction between temporary, induced dipoles	Nonpolar side-chain packing
Van der Waals radius	Effective radius of an atom's electron cloud for nonbonded contact	Spacefill sphere size; steric clashes
Steric clash (hindrance)	Atoms forced closer than their van der Waals radii allow	Forbidden Ramachandran regions; bulky residues
Hydrophobic effect	Nonpolar groups cluster away from water because ordered water around them is released, increasing solvent entropy	The main driving force of folding
Clathrate (ordered water shell)	Cage-like, more ordered water around a nonpolar surface	Its release on burial is the entropy gain of the hydrophobic effect
Dipole–dipole interaction	Attraction between permanent dipoles (e.g., aligned peptide groups)	Helix dipole; polar interactions
Ion–dipole interaction	Attraction between a charge and a polar group (e.g., Na⁺ and water, Lys⁺ and a C=O)	Hydration; helix capping
Cation–π interaction	Attraction between a cation (Lys, Arg, metal) and the electron-rich face of an aromatic ring	Binding pockets; a non-obvious noncovalent force
π-stacking	Face-to-face or edge-to-face packing of aromatic rings	Aromatic clusters; ligand binding
Disulfide bond	Covalent S–S bond between two cysteine thiols, formed by oxidation	Main covalent cross-link; extracellular proteins; HA1–HA2 link
Cystine	Two cysteines joined by a disulfide	Jmol select cystine; keratin of hair
Oxidation / reduction (thiol)	2 R–SH → R–S–S–R + 2 H⁺ + 2 e⁻ is oxidation; adding electrons breaks it (reduction)	The reducing cytosol keeps disulfides rare inside cells
Reducing agent	Reagent that breaks disulfides: dithiothreitol (DTT), β-mercaptoethanol (BME), TCEP	Reducing SDS-PAGE; separates disulfide-linked subunits
Metal coordination	Dative bonds from side-chain atoms (Cys S, His N, Asp/Glu O) to a metal ion	Zinc fingers, heme iron, Ca²⁺ in EF hands
Dielectric constant	How strongly a medium weakens electrostatic forces (water ≈ 80, protein interior ≈ 2–20, hexane ≈ 2)	Ionic interactions are much stronger when buried
Desolvation penalty	Free-energy cost of stripping water from a polar or charged group	Why burying a charge destabilizes a protein
Lennard-Jones potential	Energy model for van der Waals contact: an attractive r⁻⁶ term and a repulsive r⁻¹² term	Force fields and molecular dynamics
Force field	Set of energy terms (bond stretch, angle bend, torsions, van der Waals, electrostatics) used to compute a structure's energy	Molecular dynamics; refinement
Primary structure	The amino acid sequence, N → C	Held by peptide bonds
Secondary structure	Regular local backbone conformation (helix, strand, turn) held by backbone C=O···H–N bonds	"Which level" questions; DSSP output
Supersecondary structure (motif)	A small, recurring combination of secondary elements (β-α-β, helix-turn-helix, hairpin, Greek key)	Between secondary and tertiary in some textbooks
Tertiary structure	Full three-dimensional fold of one chain, held by side-chain interactions, the hydrophobic core and disulfides	HA1 head domain; most of what denaturation destroys
Quaternary structure	Arrangement of two or more chains (subunits)	HA homotrimer; hemoglobin α₂β₂
Quinary structure	Transient, weak contacts proteins make in the crowded cell	Distractor; not one of the four classical levels
α helix	Right-handed helix, 3.6 residues per turn, C=O(i) H-bonded to N–H(i + 4)	Most common secondary structure
3₁₀ helix	Tighter helix with 3 residues per turn and i → i + 3 H-bonds (10-atom H-bonded ring)	Short helices at helix ends; DSSP "G"
π helix	Wider helix with about 4.4 residues per turn and i → i + 5 H-bonds	Usually a π-bulge in an α helix; DSSP "I"
π-bulge	Single residue inserted into an α helix, creating a short π-helical turn of about 7 residues	Often at functional sites
Left-handed α helix (αL)	Helix with positive φ and ψ; sterically strained for L-amino acids	Single Gly or Asn residues; rarely a full helix
Polyproline II (PPII) helix	Extended left-handed helix with 3 residues per turn and no internal H-bonds	Proline-rich segments; collagen chains; unfolded proteins
Collagen triple helix	Three left-handed PPII-like chains wound into a right-handed superhelix, Gly every third residue	Fibrous protein; Gly-X-Y repeat
Helix dipole	Net dipole from aligned peptide groups: partial + at the N-terminus, − at the C-terminus	Negative residues near the N-cap, positive near the C-cap stabilize the helix
Helix capping	Side chains (often Ser, Thr, Asp, Asn at the N-cap; Gly at the C-cap) that satisfy the unpaired backbone H-bonds at helix ends	Helix start and stop signals
Helical wheel	Projection of a helix down its axis, residues 100° apart	Finding the hydrophobic face of an amphipathic helix
Heptad repeat	Seven-residue repeat abcdefg in which a and d are hydrophobic	Coiled coils and leucine zippers
Helix propensity	Tendency of a residue to adopt helical φ/ψ; Ala, Leu, Glu, Met, Gln, Lys high; Pro and Gly low	Secondary-structure prediction; design questions
Helix breaker	Residue that terminates or kinks helices, chiefly Pro (no N–H, rigid ring) and to a lesser extent Gly	"Why is Pro rare inside helices"
Helix–coil cooperativity	Nucleating the first turn is costly, but each added residue then adds one H-bond, so helix formation is cooperative (sharp transitions)	Zimm–Bragg σ (nucleation) and s (propagation) parameters
β strand	Nearly fully extended segment of backbone; side chains alternate sides	Building block of sheets; arrows in cartoons
β sheet	Two or more strands H-bonded side by side (backbone C=O to N–H)	Main secondary structure of the HA1 head
Antiparallel sheet	Adjacent strands run in opposite directions; H-bonds nearly linear	Slightly more stable than parallel; hairpins
Parallel sheet	Adjacent strands run the same way; H-bonds angled; needs long crossover connections	β-α-β units, Rossmann folds, TIM barrels
Mixed sheet	A sheet with both parallel and antiparallel strand pairs	Many α/β proteins
Pleat	Zig-zag of a sheet as successive Cα atoms go slightly up and down	Origin of "β-pleated sheet"
Sheet twist	Strands are not flat; sheets have a right-handed twist viewed along the strands	Real sheets in cartoons look twisted
β bulge	Irregularity where one strand has an extra residue opposite its partner	Breaks the alternation of side chains
Edge strand	Strand at the edge of a sheet with unpaired H-bond groups	Aggregation hot spots; protected by charges or prolines
Strand register	Which residues pair across two strands	Hairpin design; amyloid structure
Turn (reverse turn)	Short segment that reverses chain direction	Connect strands and helices; usually on the surface
β turn	Four-residue turn with C=O(i) H-bonded to N–H(i + 3); types I, II, I′, II′ and others	Gly often at position 3 of type II; Pro at position 2
γ turn	Three-residue turn with an i → i + 2 H-bond	Less common than β turns
β hairpin	Two antiparallel strands joined by a short turn	Simplest β motif; low contact order
Loop (coil)	Irregular, non-repeating backbone segment	Often flexible, surface-exposed, high B-factor
Ω loop	Loop whose ends are close together, shaped like the Greek letter omega	Surface loops in many proteins
Random coil	An ensemble of disordered conformations with no stable structure	Unfolded state; CD minimum near 198–200 nm
Torsion (dihedral) angle	Rotation about a bond, defined by four consecutive atoms	φ, ψ, ω and χ angles
φ (phi)	Backbone torsion about the N–Cα bond	Ramachandran x-axis
ψ (psi)	Backbone torsion about the Cα–C bond	Ramachandran y-axis
ω (omega)	Torsion about the peptide C–N bond, near 180° (trans) or 0° (cis)	Fixed by resonance; not a free variable
χ angles and rotamers	Side-chain torsions (χ1, χ2...) and their preferred discrete combinations (rotamers)	Side-chain modeling; alternate conformations
Ramachandran plot	Plot of ψ against φ showing sterically allowed backbone conformations	α (lower left), β (upper left), αL (upper right)
Ramachandran outlier	Residue with φ/ψ in a disallowed region	Flags errors in a structure model; Gly excepted
DSSP	Algorithm that assigns secondary structure from 3D coordinates using backbone H-bond patterns (H, G, I, E, B, T, S)	It assigns from a known structure; it does not predict from sequence
Secondary-structure prediction	Estimating helix/strand/coil per residue from sequence (Chou–Fasman propensities, GOR, neural networks trained on PDB data)	Contrast with DSSP assignment
β-α-β motif	Two parallel strands joined by a crossover helix; almost always right-handed	Rossmann fold building block
Rossmann fold	Alternating β-α-β-α-β units forming a parallel sheet that binds a nucleotide (NAD⁺, FAD)	Classic α/β fold; dehydrogenases
Helix-turn-helix (HTH)	Two helices joined by a short turn; the "recognition helix" sits in the DNA major groove	Bacterial repressors; homeodomains
Helix-loop-helix (bHLH)	Two helices joined by a longer loop, dimerizing; basic region binds DNA	Transcription factors such as Myc
EF hand	Helix–loop–helix that binds Ca²⁺ in a 12-residue loop	Calmodulin, troponin C
Zinc finger	Small domain whose fold is held by a Zn²⁺ coordinated by Cys and His (classic C2H2: ββα)	DNA binding; each classic finger reads about 3 bp
Leucine zipper (bZIP)	Coiled-coil dimerization helices with Leu at every seventh position, plus a basic DNA-binding region	Transcription factors (Fos/Jun, GCN4)
Coiled coil	Two to seven α helices wound around each other in a left-handed supercoil, held by a hydrophobic heptad seam	Keratin, leucine zippers, the HA2 stem
Greek key	Four antiparallel strands where strand 4 folds back next to strand 1	Common β topology; immunoglobulin and jelly-roll folds
β meander	Antiparallel strands each joined to its neighbor by a hairpin	Simplest sheet topology
β sandwich	Two β sheets packed face to face with a hydrophobic core between	Immunoglobulin domains
Jelly roll	β sandwich of eight strands in which hairpins are "rolled" so strand pairs interleave	Many viral capsid proteins; HA1 head
β barrel	A sheet closed on itself into a cylinder	Porins, lipocalins, GFP
TIM barrel	Eight parallel β strands surrounded by eight α helices, (βα)₈	The most common enzyme fold
β propeller	Four to eight twisted four-stranded "blades" arranged around a central axis	Influenza neuraminidase head (six blades)
Four-helix bundle	Four roughly parallel or antiparallel helices packed around a core	Cytochrome b562, growth hormone, ferritin
Immunoglobulin (Ig) fold	β sandwich of 7–9 strands in two sheets with a conserved disulfide	Antibody domains, MHC, many receptors
Flavodoxin-like fold	A central five-stranded parallel sheet flanked by helices on both sides	α/β fold example
Domain	A compact unit of a chain that can fold, and often function, on its own	HA1 head vs HA2 stem; multi-domain proteins
Fold	A specific arrangement and connectivity of secondary elements shared by many proteins	Fold libraries; threading
Topology	The order and direction of secondary elements along the chain, ignoring exact geometry	Topology diagrams of sheets
CATH / SCOP classes	Structure classifications; SCOP top classes all-α, all-β, α/β (alternating) and α+β (segregated); CATH uses mainly-α, mainly-β and α–β (both mixed kinds together)	Classifying a fold from a picture
Sequence motif	A short conserved sequence pattern linked to a function (e.g., P-loop GxxxxGK[ST], NPA in aquaporins, N-X-S/T glycosylation sequon)	Identifying function from sequence
SH2 / SH3 domain	Modules that bind phosphotyrosine peptides (SH2) or proline-rich peptides (SH3)	Signal transduction; PTM-reading domains
Transmembrane helix	A run of about 20 hydrophobic residues spanning the lipid bilayer as an α helix	Hydropathy plots; HA's membrane anchor
Positive-inside rule	Lys and Arg are enriched on the cytoplasmic side of membrane proteins	Predicting membrane topology
Hydrophobic core	The buried interior of a globular protein, packed with nonpolar side chains	Core mutations are the most destabilizing
Interface	Surface buried when two chains or a ligand meet	HA trimer interface; antibody footprints
Subunit (protomer)	One chain, or the repeating unit, of a multi-chain protein	HA protomer = HA1 + HA2
Homo- / hetero-oligomer	Assembly of identical (homo) or different (hetero) subunits	HA is a homotrimer; hemoglobin a heterotetramer
Symmetry (Cₙ)	n-fold rotational symmetry of an oligomer	HA trimer has 3-fold (C3) symmetry
Solvent-accessible surface area (SASA)	Area traced by the center of a water-sized probe rolled over the protein	Buried vs exposed residues; hydrophobic effect estimates
Packing density	Fraction of space filled by atoms; protein interiors are about 0.7–0.75, like a packed solid	"Proteins are well packed, not hollow"
Contact order	Average sequence separation of residues in contact in the native state, normalized for length	Low contact order (hairpins, helices) → faster folding
Native state	The folded, functional conformation	Folding questions; "native conformation"
Denatured (unfolded) state	Ensemble of non-native, largely disordered conformations	Loss of function; still has intact peptide bonds
Denaturation	Loss of secondary, tertiary and quaternary structure without breaking peptide bonds	Heat, pH extremes, urea, detergents, organic solvents
Renaturation	Spontaneous refolding of a denatured protein when the denaturant is removed	Anfinsen's ribonuclease experiment
Anfinsen's dogma (thermodynamic hypothesis)	The native fold is the global free-energy minimum and is encoded in the sequence; it needs a unique, stable and kinetically accessible minimum	Sequence determines structure; prions are the famous exception
Levinthal's paradox	A chain could never find its fold by random search of all conformations, yet real proteins fold in microseconds to seconds	Folding is guided, not random
Folding funnel (energy landscape)	Picture of folding as downhill flow on a rugged, funnel-shaped free-energy surface toward the native state	Explains multiple routes and kinetic traps
Principle of minimal frustration	Evolved sequences have native contacts that are mutually consistent, giving smooth, funneled landscapes (Bryngelson and Wolynes)	Named-principle questions
Kinetic trap	A local free-energy minimum where a misfolded chain can get stuck	Misfolding; why chaperones help
Molten globule	Compact intermediate with native-like secondary structure but loosely packed side chains	Folding intermediates
Hydrophobic collapse	Early, rapid compaction of the chain driven by burying nonpolar groups	Folding mechanism models
Nucleation–condensation	Model in which a small folding nucleus forms and the rest of the structure condenses around it	Folding mechanism models
Framework model	Model in which secondary structure forms first and then docks into tertiary structure	Folding mechanism models
Folding transition state	The highest free-energy point on the folding pathway that every folding molecule must pass	Saddle point of the landscape; sets the rate
Two-state folding	Only native and unfolded states are significantly populated	Basis of ΔG and T_m calculations
Gibbs free energy (ΔG)	$\Delta G = \Delta H - T\Delta S$; negative means spontaneous	Folding has ΔG < 0
Enthalpy (ΔH)	Heat change, mostly from bonds and interactions made or broken	Folding enthalpy from H-bonds and van der Waals contacts
Entropy (ΔS)	Measure of disorder (number of accessible microstates)	Chain loses entropy on folding; water gains it
Conformational entropy	Entropy of the chain's own flexibility, lost on folding	The main force opposing folding
Melting temperature (T_m)	Temperature at which half the molecules are unfolded (ΔG_unfold = 0)	Stability comparisons; DSC and CD melts
ΔC_p (heat-capacity change)	Increase in heat capacity on unfolding, from newly exposed nonpolar surface	Curved stability vs temperature; cold denaturation
Cold denaturation	Unfolding at low temperature, predicted by the curved stability curve	Stability is maximal at an intermediate temperature
Marginal stability	Native proteins are only about 20–60 kJ/mol more stable than unfolded ones	Why one mutation or a fever can matter
Chaotrope	Solute that disrupts water structure and weakens the hydrophobic effect (urea, guanidinium, SCN⁻)	Chemical denaturation
Kosmotrope	Solute that strengthens water structure and stabilizes folded proteins (sulfate, phosphate)	Hofmeister series; salting out
Hofmeister series	Ranking of ions by how they stabilize (kosmotropes such as sulfate) or destabilize (chaotropes such as thiocyanate) proteins	Salt effects on stability and solubility
Salting out	Precipitating proteins at high salt (e.g., ammonium sulfate)	Purification and crystallization
Metastable protein	Protein held in a higher-energy state by a kinetic barrier and released by a trigger	HA's pre-fusion form; serpins
Cotranslational folding	Folding of the nascent chain while it is still being made on the ribosome	Why in-vitro refolding can be slower
Molecular chaperone	Protein that binds exposed hydrophobic surfaces of non-native chains to prevent aggregation (e.g., Hsp70)	Folding helpers; does not encode the fold
Chaperonin	Barrel-shaped chaperone that encloses a chain in a cavity to fold (GroEL/GroES in bacteria, TRiC/CCT in eukaryotes)	Chaperone vs chaperonin questions
Heat shock protein (Hsp)	Chaperones induced by heat or stress, named by mass in kDa (Hsp60, Hsp70, Hsp90)	Stress response
Protein disulfide isomerase (PDI)	ER enzyme that forms and shuffles disulfides	Correct disulfide pairing
Peptidyl-prolyl isomerase	Enzyme that speeds cis–trans isomerization of X–Pro bonds	Slow folding steps
Unfolded protein response	ER stress signaling that boosts chaperones and slows translation when misfolded proteins build up	Cell biology of misfolding
Ubiquitin–proteasome system	Ubiquitin chains (Lys48-linked) tag proteins for degradation by the 26S proteasome	Disposal of misfolded proteins
Aggregation	Nonspecific association of misfolded chains, often via exposed hydrophobic surfaces	Loss of function; disease
Amyloid	Insoluble fibrils with a cross-β structure	Alzheimer's (Aβ, tau), Parkinson's (α-synuclein), type 2 diabetes (amylin)
Cross-β structure	Strands run perpendicular to the fibril axis, sheets parallel to it	Diagnostic X-ray reflections at 4.7 Å and about 10 Å
Prion	Misfolded protein (PrPˢᶜ) that converts the normal form (PrPᶜ) into more of itself	CJD, kuru, BSE; infectious protein without nucleic acid
Intrinsically disordered protein (IDP)	Protein or region with no stable fold under physiological conditions, often rich in charged and polar residues	Exceptions to "structure determines function"; tau
Hydrogen–deuterium exchange	Exchange of amide H for D in D₂O, slowed where the H is H-bonded or buried	Probing folding and protection (NMR, mass spec)
Enzyme	Biological catalyst (usually protein, sometimes RNA) that speeds a reaction without being consumed	Function questions
Ribozyme	Catalytic RNA (e.g., the ribosome's peptidyl transferase center)	"Enzymes can also be made of RNA"
Active site	Pocket or cleft where substrate binds and catalysis occurs	Conserved residues; inhibitor target
Substrate	Molecule an enzyme acts on	Specificity questions
Activation energy (Eₐ)	Free-energy barrier between reactants and the transition state	Enzymes lower it; they do not change ΔG of the reaction
Transition-state stabilization	Enzymes bind the transition state more tightly than the substrate	Core of catalysis; transition-state analog inhibitors
Catalytic triad	Ser–His–Asp (or Cys–His) arrangement making a strong nucleophile in proteases	Serine and cysteine proteases
Lock and key	Model in which substrate fits a rigid, pre-formed site (Fischer)	Contrast with induced fit
Induced fit	Model in which binding reshapes the site to fit the substrate (Koshland)	Conformational change on binding
Conformational selection	Ligand selects and stabilizes a pre-existing conformation from an ensemble	Modern view alongside induced fit
Cofactor	Non-protein helper an enzyme needs: a metal ion or an organic coenzyme	Zn²⁺, Fe²⁺, Mg²⁺; NAD⁺, FAD
Coenzyme	Organic cofactor, often vitamin-derived (NAD⁺ from niacin, FAD from riboflavin)	Cofactor classification
Prosthetic group	Cofactor bound tightly or covalently (heme in hemoglobin)	Cofactor classification
Apoenzyme / holoenzyme	Enzyme without / with its cofactor	Vocabulary
Ligand	Any molecule or ion that binds specifically to a protein	Jmol select ligand; LSTc in 3UBE
Binding pocket	Region whose shape and chemistry complement a ligand	Receptor binding site of HA
Affinity	Strength of a single binding interaction, measured by the dissociation constant K_d (lower K_d = tighter)	Antibody and receptor binding
Avidity	Combined strength of several simultaneous binding interactions, far greater than any single one	IgM's 10 sites; trimeric HA binding many sialic acids
Specificity	Preference for one partner over others, from shape and chemical complementarity	Enzymes, antibodies, HA linkage preference
Competitive inhibition	Inhibitor binds the active site and competes with substrate	Raises apparent K_M; overcome by more substrate (oseltamivir)
Noncompetitive inhibition	Inhibitor binds elsewhere equally well to E and ES, lowering activity	Lowers V_max
Uncompetitive inhibition	Inhibitor binds only the enzyme–substrate complex	Lowers both V_max and K_M
Mixed inhibition	Inhibitor binds E and ES with different affinities	General case of the above
Allosteric regulation	Binding at one site changes activity at a distant site via a conformational change	Hemoglobin O₂ cooperativity; feedback inhibition
Cooperativity	Binding at one site changes affinity at others (positive in hemoglobin)	Sigmoidal binding curve
Irreversible (covalent) inhibitor	Inhibitor that bonds covalently to the enzyme, inactivating it permanently	Aspirin acetylates COX; penicillin
Feedback inhibition	A pathway's end product inhibits an early enzyme	Regulation questions
Zymogen (proenzyme)	Inactive precursor activated by cleavage (trypsinogen, pepsinogen)	Prevents self-digestion; compare HA0 activation
Diffusion limit	Maximum rate set by how fast substrate reaches enzyme, about 10⁸–10⁹ M⁻¹ s⁻¹	"Catalytically perfect" enzymes
Electrostatic steering	Charged residues guide a charged substrate into the active site, letting rates approach or exceed simple diffusion estimates	Superoxide dismutase, acetylcholinesterase
Enzyme Commission (EC) classes	Seven classes: oxidoreductases, transferases, hydrolases, lyases, isomerases, ligases, translocases	Classifying a reaction
Protease	Enzyme that hydrolyzes peptide bonds (serine, cysteine, aspartic, metallo-)	HA0 cleavage by host trypsin-like proteases
Kinase / phosphatase	Enzymes that add / remove phosphate groups	Signaling; phosphorylation questions
Nuclease	Enzyme that cleaves DNA or RNA	PA endonuclease of influenza
Isozyme	Different forms of an enzyme (distinct sequences, often from different genes) catalyzing the same reaction	Vocabulary
Globular protein	Compact, roughly spherical, water-soluble protein with a hydrophobic core	Enzymes, antibodies, HA1 head
Fibrous protein	Elongated protein with repetitive structure, usually structural and insoluble	Collagen, keratin, silk fibroin
Membrane protein	Protein embedded in (integral) or attached to the surface of (peripheral) a lipid bilayer	Hydrophobic outer surface in the bilayer
Type I membrane protein	Single-pass membrane protein with its N-terminus outside the cell and C-terminus in the cytoplasm	HA is type I; NA is type II (N-terminal anchor, C-terminus outside); M2 is N-out like HA but has no cleaved signal peptide, so it is usually called type III
GPCR	G-protein-coupled receptor with 7 transmembrane helices	Rhodopsin, β-adrenergic receptor
Rhodopsin	GPCR in rod cells whose bound retinal isomerizes on absorbing light, activating the G protein transducin	Light → signal conversion
Ion channel	Membrane protein forming a gated pore for ions	Faster than GPCR signaling; M2 proton channel
Transporter	Membrane protein moving solutes by conformational change (uniporter, symporter, pump)	Vocabulary
Receptor	Protein that binds a signal (hormone, ligand) and triggers a response	Function categories
Transcription factor	Protein that binds DNA to regulate transcription	Zinc fingers, HTH, bZIP
Structural protein	Protein that provides support or shape	Collagen, keratin, actin
Motor protein	Protein that converts ATP energy into movement (myosin, kinesin, dynein)	Function categories
Collagen	Triple-helical fibrous protein of connective tissue, Gly-X-Y repeat, Pro/hydroxyproline-rich	Most abundant protein in mammals; scurvy
α-Keratin	Coiled-coil fibrous protein of hair and nails, cross-linked by disulfides	Perms break and reform disulfides
Silk fibroin	Stacked antiparallel β sheets of Gly-Ala-Gly-Ala-Gly-Ser repeats	Strong, flexible β-sheet fiber
Hemoglobin / myoglobin	O₂-binding heme proteins; myoglobin is one chain, hemoglobin α₂β₂ and cooperative	Allostery, Bohr effect, sickle cell
Bohr effect	Lower pH and higher CO₂ reduce hemoglobin's O₂ affinity, partly via His protonation and new salt bridges	pKₐ and salt-bridge questions
Heme	Protoporphyrin ring with a central Fe²⁺, bound to the proximal His	Prosthetic group
Lectin	Protein that binds specific carbohydrates	HA is a viral lectin for sialic acid
Glycoprotein	Protein carrying covalently attached sugars (glycans)	HA and NA; most secreted proteins
Post-translational modification (PTM)	Covalent change to a protein after translation	Expands chemistry beyond 20 amino acids
Phosphorylation	Addition of phosphate to Ser, Thr or Tyr (His in bacteria) by kinases	Signaling switches; about a third of human proteins
N-linked glycosylation	Glycan attached to the amide N of Asn in an Asn-X-Ser/Thr sequon (X not Pro), starting in the ER	HA glycan shield; masking antigenic sites
O-linked glycosylation	Glycan attached to the O of Ser or Thr, added in the Golgi	Mucins
Sequon	The N-X-S/T consensus that allows N-glycosylation	Glycan sites gained in antigenic drift
Acetylation	Addition of an acetyl group to the N-terminus or a Lys ε-amino group	Lys loses its + charge; histone acetylation opens chromatin
N-terminal acetylation	Acetyl added to the α-amino group, on most human proteins	Affects stability and degradation signals
Methylation	Addition of methyl groups to Lys or Arg (proteins) or cytosine (DNA)	Histone code; gene silencing
Ubiquitination	Attachment of ubiquitin to Lys via an isopeptide bond	Lys48 chains → proteasome; Lys63 chains → signaling
SUMOylation	Attachment of the small ubiquitin-like modifier to Lys	Nuclear transport; regulation
Palmitoylation (S-acylation)	Reversible attachment of palmitate (C16) to Cys by a thioester	Membrane anchoring; HA's cytoplasmic tail cysteines
Myristoylation	Attachment of myristate (C14) to an N-terminal Gly	Membrane anchoring
Prenylation	Attachment of a farnesyl or geranylgeranyl group to a C-terminal Cys	Ras membrane targeting
Hydroxylation	Addition of –OH, e.g., Pro → hydroxyproline in collagen (vitamin C-dependent)	Scurvy; collagen stability
γ-Carboxylation	Vitamin K-dependent conversion of Glu to γ-carboxyglutamate (Gla) for Ca²⁺ binding	Blood-clotting factors
GPI anchor	Glycolipid anchoring a protein's C-terminus to the outer leaflet of the membrane	Prion protein
Proteolytic processing	Activating or maturing a protein by cutting its chain	HA0 → HA1 + HA2; zymogens; insulin
Signal peptide	N-terminal sequence (hydrophobic core of about 7–15 residues) directing a protein into the ER	HA's 17-residue signal peptide is cleaved off
Signal recognition particle (SRP)	Ribonucleoprotein that binds a signal peptide and docks the ribosome on the ER	Secretory pathway
Codon	Three-nucleotide mRNA unit specifying an amino acid or stop	Translation questions
Anticodon	Complementary three-base sequence on a tRNA	Base pairing with codons
Start codon	AUG (Met), which also sets the reading frame	Why Met is first
Stop codon	UAA, UAG, UGA, which code for no amino acid in the standard code	Nonsense mutations
Degeneracy (redundancy)	Most amino acids have more than one codon	Silent mutations
Wobble	Non-standard pairing at the codon's third base lets one tRNA read several codons	It does not let one mRNA make different proteins
Reading frame / ORF	The grouping of bases into codons; an open reading frame runs from a start to a stop codon	Frameshifts
Exon / intron	Coding segments kept in mature mRNA / non-coding segments spliced out	Splicing mutations; influenza M and NS mRNAs are spliced
Exome	All exons of a genome; about 1–2% of human DNA	Vocabulary
Point mutation	A change of a single base	Substitution categories
Missense mutation	Point mutation that changes one amino acid	HA drift; sickle cell E6V
Nonsense mutation	Point mutation that creates a premature stop codon	Truncated protein
Silent (synonymous) mutation	Base change that leaves the amino acid unchanged	Usually third codon position
Frameshift	Insertion or deletion not a multiple of three, shifting all downstream codons	Usually a garbled, truncated protein
In-frame indel	Insertion or deletion of a multiple of three bases	CFTR ΔF508 deletes one Phe
Transition / transversion	Purine↔purine or pyrimidine↔pyrimidine change / purine↔pyrimidine change	C → T is a transition
Conservative substitution	Replacement by a chemically similar residue (Leu → Ile, Lys → Arg, Asp → Glu)	Usually tolerated
Nonconservative substitution	Replacement by a dissimilar residue (charge, size or polarity change)	Core Ile → Pro, Glu → Val
Mutation notation	Wild-type residue, position, new residue (D225G = Asp225 → Gly)	Reading HA and SARS-CoV-2 variant names
Loss / gain of function	Mutation reducing / adding activity; dominant negative = a mutant that blocks the normal copy	Effect-prediction questions
Site-directed mutagenesis	Deliberately changing chosen codons in a cloned gene	Testing the role of a residue
Alanine scanning	Mutating residues one by one to Ala to map which matter	Binding hot spots
Conserved residue	Residue unchanged across related sequences, usually because it is structurally or functionally essential	Active sites, receptor-binding-site residues
Sequence alignment	Lining up sequences to match equivalent residues, allowing gaps	Conservation; homology modeling
Identity vs similarity	Percent identical residues vs percent identical-or-conservatively-substituted	Homology thresholds
Substitution matrix	Scores for replacing one residue by another, from observed alignments (BLOSUM, PAM)	Scoring alignments
Homolog / ortholog / paralog	Shared ancestry / in different species (speciation) / in one genome (duplication)	Evolution vocabulary
Consensus sequence	The most common residue at each position of an alignment	Motifs; vaccine design
X-ray crystallography	Structure determination from the diffraction of X-rays by a protein crystal	About 80% of PDB entries; 3UBE method
Bragg's law	$n\lambda = 2d\sin\theta$: condition for constructive reflection from lattice planes spaced d	Plane-spacing calculations
Unit cell	Smallest box that, repeated by translation, builds the crystal	Cell edges a, b, c and angles α, β, γ
Space group	The set of symmetry operations of the crystal (65 possible for chiral molecules such as proteins)	Crystallography vocabulary
Asymmetric unit	Smallest part of the crystal from which symmetry generates the whole unit cell; the coordinates actually deposited	May contain several copies or only part of the biological assembly
Biological assembly	The functional oligomer, which may differ from the asymmetric unit	HA trimer generated from chains in the file
Reflection	One diffraction spot, indexed by Miller indices (hkl)	Data collection
Structure factor	Wave scattered for one reflection, with an amplitude (measured) and a phase (not measured)	Phase problem
Phase problem	Detectors record intensities but not phases, which are needed to compute the electron density	Solved by molecular replacement or experimental phasing
Molecular replacement	Getting initial phases by placing a known similar structure in the new crystal	3UBE was solved this way
Experimental phasing	Phases from heavy-atom or anomalous scatterers (MIR, SAD, MAD; often selenomethionine)	For structures with no similar model
Electron density map	3D map of electron density computed from amplitudes and phases, into which the model is built	Mesh displays; quality at a given resolution
Resolution	The smallest plane spacing with measurable data; smaller numbers mean finer detail	Judging a structure
R-factor (R-work)	Disagreement between observed and model-calculated amplitudes; lower is better	Model quality
R-free	R computed on 5–10% of reflections excluded from refinement	Guards against overfitting
B-factor (temperature factor)	Measure of each atom's positional spread, $B = 8\pi^2\langle u^2\rangle$	High in loops and surface side chains; color temperature
Occupancy	Fraction of molecules in which an atom is at that position (1.0 = always)	Alternate conformations; partial ligands
Alternate conformation	Two or more modeled positions for an atom or side chain (altloc A, B)	Flexible side chains
Matthews coefficient (V_M)	Crystal volume per dalton of protein in the unit cell	Estimates solvent content and copies per asymmetric unit
Vapor diffusion	Crystallization by slowly drawing water from a hanging or sitting drop toward a reservoir	Crystal growth; first step is nucleation
Synchrotron	Particle accelerator producing intense, tunable X-rays	Most modern data collection
Cryo-cooling	Flash-freezing crystals (about 100 K) to reduce radiation damage	Data collection
Structure validation	Checking a model's geometry and fit (Ramachandran outliers, clashscore, R-free)	wwPDB validation reports
NMR spectroscopy	Structure from nuclear spin signals of a protein in solution	Small proteins; dynamics; ensembles of models
Chemical shift	Resonance frequency of a nucleus relative to a reference (ppm), set by its environment	Folded vs unfolded spectra
NOE	Nuclear Overhauser effect: signal between nuclei closer than about 5–6 Å	Distance restraints for NMR structures
NMR ensemble	A set of models (often 10–20) all consistent with the data	Spread shows flexibility
Cryo-electron microscopy (cryo-EM)	Structure from many electron images of frozen, randomly oriented particles	Large complexes; membrane proteins
Vitreous ice	Non-crystalline ice formed by plunge-freezing very fast	Preserves particles for cryo-EM
Particle picking / 2D class averages	Locating particles in micrographs / averaging particles in similar orientations	Cryo-EM workflow steps
3D reconstruction	Combining 2D views of known orientation into a 3D map	Cryo-EM workflow
Fourier shell correlation (FSC)	Agreement between two half-maps vs resolution; cryo-EM resolution is quoted where FSC = 0.143	Cryo-EM resolution
Protein Data Bank (PDB)	The single global archive of experimentally determined 3D structures of macromolecules (wwPDB; RCSB in the US)	Source of 3UBE
PDB ID	Four-character identifier, digit first (e.g., 3UBE)	Loading a structure in JUDE
PDB / mmCIF file	Text files holding coordinates and metadata; mmCIF is the current standard format	What a structure file contains
ATOM / HETATM records	Lines for standard polymer atoms / for ligands, waters, ions and modified residues	Selecting ligands and water
Chain ID	Letter naming each chain in a file	select 190:A; 3UBE chain A = HA1
RMSD	Root-mean-square deviation of matched atoms after superposition, in Å	Comparing structures or predictions
Superposition	Rotating and translating one structure onto another to minimize RMSD	Structure comparison
Homology (comparative) modeling	Building a model on the structure of a related template	Works well above about 30% identity
Template	Experimentally solved structure used to build a model	Homology modeling
Threading (fold recognition)	Testing a sequence against a library of known folds	Remote homologs
Ab initio (de novo) prediction	Predicting structure from physics or statistics without a template	Rosetta-style methods
AlphaFold	DeepMind's deep-learning structure predictor; AlphaFold2 earned the 2024 Chemistry Nobel (Hassabis, Jumper; shared with Baker for design)	Prediction questions
pLDDT	AlphaFold's per-residue confidence score (0–100)	Low scores often mark disordered regions
PAE	Predicted aligned error: AlphaFold's estimate (Å) of relative position error between residue pairs	Domain placement confidence
Molecular dynamics	Simulation of atomic motion over time using a force field	Flexibility; folding simulations
De novo protein design	Designing new sequences that fold into chosen structures ("designer proteins")	Baker lab; Nobel 2024
Edman degradation	Stepwise N-terminal sequencing: PITC labels the N-terminal residue, acid cleaves it off, and it is identified as a PTH-amino acid	One residue per cycle; blocked by N-terminal acetylation
Phenyl isothiocyanate (PITC)	Edman's reagent; couples to the free α-amino group at mildly alkaline pH	Edman chemistry
Sanger's reagent (FDNB)	1-Fluoro-2,4-dinitrobenzene, labels the N-terminal residue (destroying the rest of the chain on hydrolysis)	Insulin, the first protein sequenced
Mass spectrometry	Measures mass-to-charge ratio of ionized molecules (ESI, MALDI); tandem MS sequences peptides	Protein ID; PTM mapping
Peptide mass fingerprinting	Identifying a protein from the masses of its tryptic peptides	Proteomics
SDS-PAGE	Electrophoresis of SDS-denatured proteins, separating by size (mass)	Subunit mass; reducing vs non-reducing gels reveal disulfide-linked subunits
Native PAGE	Electrophoresis without denaturant; separates by size, shape and charge	Oligomeric state
Western blot	Gel-separated proteins transferred to a membrane and detected with antibodies	Detecting a specific protein
Isoelectric focusing (IEF)	Electrophoresis in a pH gradient; each protein stops at its pI	First dimension of 2D gels
2D gel electrophoresis	IEF (by pI) then SDS-PAGE (by mass) at right angles	Proteome mapping
Size-exclusion chromatography	Separation by size; large molecules elute first because they skip the bead pores	Oligomeric state; gel filtration
Ion-exchange chromatography	Separation by charge on a charged resin	Purification; depends on pH vs pI
Affinity chromatography	Purification by specific binding to an immobilized partner	His-tag, antibody columns
His-tag	Run of about six His residues added to a protein so it binds Ni²⁺ (or Co²⁺) resin; eluted with imidazole	Recombinant purification
Circular dichroism (CD)	Differential absorption of left vs right circularly polarized UV light, reporting secondary structure	Helix minima at 208 and 222 nm; thermal melts
Fluorescence spectroscopy	Emission after excitation; Trp emission shifts when Trp becomes exposed on unfolding	Folding studies
UV absorbance (A₂₈₀)	Absorbance at 280 nm, mainly from Trp and Tyr	Measuring protein concentration
Differential scanning calorimetry (DSC)	Measures heat absorbed while heating, giving T_m and ΔH of unfolding	Stability
Isothermal titration calorimetry (ITC)	Measures heat released or absorbed as a ligand is titrated in, giving K_d, ΔH and stoichiometry	Binding thermodynamics
Surface plasmon resonance (SPR)	Label-free, real-time binding on a sensor surface, giving on and off rates and K_d	Antibody and receptor kinetics
Dialysis	Exchanging buffer through a semipermeable membrane	Removing denaturant to refold
Recombinant expression	Producing a protein from a cloned gene in a host (E. coli, yeast, insect or mammalian cells)	Proteins from other species may misfold
JUDE	3D Molecular Designs' browser-based Jmol viewer used in the event	Exam visualization tasks
Jmol / JSmol	Open-source molecule viewer (JSmol is the JavaScript version) driven by commands	Command questions
Representation	The drawing style of the molecule	Pros and cons of each style
Wireframe	Every bond drawn as a thin line	All atoms visible, cluttered
Sticks	Bonds drawn as cylinders	Side chains and ligands
Ball and stick	Small atom spheres joined by stick bonds	Measuring and showing pockets
Spacefill (CPK)	Atoms as van der Waals spheres	True shape and packing; hides interior
Cartoon / ribbon	Helices as spirals or cylinders, strands as arrows (N → C)	Secondary structure at a glance
Trace / backbone	A line or tube through the α-carbons	Matches the physical model
Molecular surface	Surface touched by a rolling probe (Connolly surface)	Pockets and grooves
CPK coloring	Element colors: C gray, O red, N blue, S yellow, H white, P orange	color cpk
Color structure	Helix magenta, strand yellow, turn blue, coil white	color structure
Color temperature	Coloring by B-factor, blue low to red high	Finding flexible regions
Color group	Rainbow along the chain, blue at the N-terminus to red at the C-terminus	Following chain direction
Selection expression	Jmol syntax that picks atoms (residue numbers, chain, sets, and/or/not, within)	Every command acts on the current selection
Restrict	Jmol command that selects a set and hides everything else	Showing only 153–252:A
Water (HOH)	Crystallographic water molecules listed as HETATM	select water, hide water
Alpha-carbon backbone model	Physical model showing one point per residue joined at 3.8 Å spacing	The 2027 model type
Legend	The model's printed key explaining what, how and why each feature is shown	Scored with the model
Virus	Obligate intracellular parasite: nucleic acid in a protein coat, sometimes with a lipid envelope	Not cellular; cannot replicate alone
Virion	One complete, infectious virus particle	Counting particles; morphology
Capsid	The protein shell surrounding the genome	Naked vs enveloped viruses
Capsomere	Protein subunit cluster from which a capsid is built	Icosahedral capsids
Nucleocapsid	Genome plus its tightly bound capsid or nucleoprotein	Influenza vRNP (helical)
Envelope	Lipid bilayer taken from host membrane, studded with viral glycoproteins	Influenza is enveloped; sensitive to soap and alcohol
Spike (envelope glycoprotein)	Surface protein projecting from the envelope	HA and NA spikes
Matrix protein	Protein lining the inside of the envelope, linking it to the nucleocapsid	Influenza M1
Tropism	Which hosts, tissues or cells a virus can infect, often set by receptor distribution	Upper vs lower airway infection
Host range	The set of species a virus can infect	Avian vs human influenza
Viral receptor	Host molecule the virus attaches to	Sialic acid for influenza HA
Sialic acid	Nine-carbon acidic sugar (Neu5Ac in humans) at the tips of cell-surface glycans	HA receptor; NA substrate
α2,3 / α2,6 linkage	How sialic acid joins the next galactose; birds' gut has α2,3, the human upper airway α2,6	Host specificity of HA
Receptor binding site (RBS)	Shallow pocket at the top of the HA1 head formed by the 130-loop, 150-loop, 190-helix and 220-loop	The 150-loop, 190-helix and 220-loop lie inside the 153–252 model segment; the 130-loop (about 133–138, H3 numbering as in 3UBE) lies just before it
Endocytosis	Uptake of a bound virion into an endosome	Influenza entry route
Membrane fusion	Merging of viral and host membranes to release the genome	HA2-driven, triggered by low pH
Fusion peptide	Hydrophobic N-terminal segment of HA2 that inserts into the target membrane	Exposed by HA0 cleavage
Spring-loaded mechanism	Low-pH refolding of HA2 in which a loop becomes helix, firing the fusion peptide toward the target membrane	Metastable proteins
HA0 cleavage	Host protease cuts the HA0 precursor into disulfide-linked HA1 and HA2, priming it for fusion	Required for infectivity
Polybasic cleavage site	Multiple Arg/Lys at the HA0 cleavage site, cut by ubiquitous furin-like proteases	Highly pathogenic avian H5 and H7
Uncoating	Release of the genome from the capsid or vRNP into the cell	M2 acidifies the virion interior to help it
Negative-sense RNA	Genome complementary to mRNA, which must be copied by a viral polymerase before translation	Influenza (Baltimore Group V)
Segmented genome	Genome split into separate molecules (eight for influenza A and B)	Enables reassortment
RNA-dependent RNA polymerase (RdRp)	Viral enzyme copying RNA from an RNA template, without proofreading	PB1, PB2, PA heterotrimer
Cap-snatching	Influenza polymerase steals 5′ caps from host mRNAs to prime its own transcription	PB2 binds caps; PA cuts them
vRNP	Viral ribonucleoprotein: one RNA segment coated with NP plus the polymerase	Unit packaged into virions
Budding	Release of enveloped virions by wrapping in host membrane	Influenza buds from the apical plasma membrane
Reassortment	Swapping of whole segments between two viruses infecting the same cell	Antigenic shift; 2009 pandemic origin
Antigenic drift	Gradual accumulation of point mutations in antigenic sites	Annual vaccine updates
Antigenic shift	Abrupt acquisition of a new HA (± NA) subtype, by reassortment or direct jump	Pandemics
Subtype	Influenza A classification by HA and NA (H1–H18, N1–N11)	Strain names such as H1N1
Clade / lineage	A group of related viruses descended from a common ancestor	Vaccine strain selection; B/Victoria and B/Yamagata
Quasispecies	Swarm of closely related variant genomes within one host	Error-prone RdRp
Zoonosis (spillover)	Infection that passes from animals to humans	Avian and swine influenza
Reservoir	The host population in which a pathogen persists	Wild aquatic birds for influenza A
Mixing vessel	Host infectable by both avian and human viruses, allowing reassortment	Classically the pig
Epidemic / pandemic / endemic	Unusual rise in one region / worldwide spread of a new strain / constant baseline presence	Vocabulary
Baltimore classification	Seven virus groups by genome type and mRNA route	Influenza is Group V, (−)ssRNA
Lytic / lysogenic cycle	Phage replicates and bursts the cell / phage genome integrates and is carried silently	Virus-cycle vocabulary
Antigenic site	Region on a protein surface recognized by antibodies; H1 sites are Sa, Sb, Ca1, Ca2 and Cb (H3: A–E)	Mostly on the HA1 head near the RBS
Glycan shield	Glycans covering a protein surface that hide epitopes from antibodies	HA glycosylation sites
HA stem (stalk)	Membrane-proximal HA region, mostly HA2, more conserved than the head	Target of broadly neutralizing antibodies and universal-vaccine designs
Neuraminidase inhibitor	Antiviral that blocks NA so progeny stay stuck to cells (oseltamivir, zanamivir, peramivir)	Treatment questions
Antigen	Any molecule recognized by an antibody or T-cell receptor	Immunology vocabulary
Immunogen	Antigen that actually provokes an immune response	Vaccine design
Epitope	The specific part of an antigen contacted by an antibody or T-cell receptor	HA antigenic sites
Linear vs conformational epitope	Continuous stretch of sequence vs residues brought together only by folding	Most HA-neutralizing epitopes are conformational
Paratope	The antigen-binding surface of an antibody	Formed by the CDR loops
Antibody (immunoglobulin)	Y-shaped protein of two heavy and two light chains joined by disulfides	Neutralization
Fab / Fc	Antigen-binding arm / constant stem that sets class and effector functions	Antibody structure
CDR	Complementarity-determining region: three hypervariable loops per chain, six per binding site	Antibody specificity
Isotype (class)	Antibody class set by the heavy-chain constant region: IgM, IgG, IgA, IgE, IgD	Function of each class
Class switching	B cell changes its antibody class (IgM → IgG/IgA/IgE) while keeping specificity	Requires T-cell help
Affinity maturation	Somatic hypermutation plus selection in germinal centers raises antibody affinity	Better antibodies over time
Neutralization	Antibody binding that prevents infection (blocking attachment, fusion or release)	Main protective mechanism against influenza
Broadly neutralizing antibody	Antibody that neutralizes many strains or subtypes, often by binding the conserved HA stem	Universal-vaccine research
Titer	Highest dilution of serum still showing activity, written as a reciprocal (1:40 = 40)	HAI and neutralization assays
Hemagglutination	Clumping of red blood cells by virus binding sialic acid on several cells	Basis of HA assays; gave HA its name
HAI (hemagglutination inhibition) assay	Measures antibodies that stop virus from agglutinating red blood cells	Standard flu-immunity measure
Seroconversion	A significant rise in specific antibody, for HAI usually a fourfold or greater increase	Vaccine trial endpoint
Seroprotection	Antibody level linked to protection, conventionally HAI titer ≥ 40	Vaccine trial endpoint
Memory cell	Long-lived B or T cell that makes a faster, stronger secondary response	Basis of vaccination
Original antigenic sin (imprinting)	Lifelong bias of the antibody response toward the first influenza strains a person met	Birth-year effects on susceptibility
Innate immunity	Fast, non-specific first-line defense (barriers, interferon, phagocytes, NK cells, complement)	First hours to days of infection
Adaptive immunity	Slower, antigen-specific response with memory (B and T cells)	Vaccines rely on it
Pattern-recognition receptor (PRR)	Sensor of pathogen-associated molecular patterns (PAMPs), e.g., TLR3, TLR7, RIG-I	Innate detection of viral RNA
Interferon	Cytokine signal that puts nearby cells into an antiviral state (type I α/β, type II γ, type III λ)	Countered by influenza NS1
Cytokine	Small signaling protein of the immune system (interleukins, TNF, interferons)	Fever, inflammation
Cytokine storm	Excessive, damaging cytokine release	Severe 1918 and H5N1 disease
MHC class I / II	Molecules that present peptides from inside the cell to CD8⁺ T cells / from engulfed antigen to CD4⁺ T cells	Rule of 8: CD8 × I, CD4 × II
HLA	Human leukocyte antigen, the human MHC (class I: HLA-A, -B, -C; class II: DP, DQ, DR)	Transplant and T-cell vocabulary
Antigen-presenting cell (APC)	Dendritic cells, macrophages and B cells that present antigen on MHC II	Activating helper T cells
CD4⁺ helper T cell	T cell that recognizes MHC II and helps B cells and CD8⁺ cells	Class switching, affinity maturation
CD8⁺ cytotoxic T cell	T cell that recognizes MHC I and kills infected cells (perforin, granzymes)	Cross-reactive anti-NP immunity
NK cell	Innate killer that attacks cells with low MHC I or antibody coating	Missing-self recognition; ADCC
Complement	About 30 plasma proteins forming a cascade that tags or lyses pathogens	Classical, lectin and alternative pathways
ADCC	Antibody-dependent cellular cytotoxicity: NK cells kill antibody-coated infected cells via Fc receptors	Non-neutralizing antibody protection
Adjuvant	Vaccine ingredient that boosts the immune response (aluminum salts, MF59, AS03)	Vaccine questions
Inactivated vaccine	Killed or split virus, or purified HA, given by injection	Most flu shots
Live attenuated vaccine (LAIV)	Weakened, cold-adapted virus given as a nasal spray	Mucosal IgA and T-cell responses
Recombinant HA vaccine	HA made in insect cells without eggs	Egg-free vaccine
Herd immunity	Indirect protection when enough of a population is immune that transmission falters	Public-health vocabulary
Secretory IgA	Dimeric IgA secreted onto mucosal surfaces	First antibody barrier in the airway

The glossary defines; the earlier chapters explain. If a definition here prompts a "why", the section named in the third column has the mechanism and the pitfalls.

Numbers worth memorizing

Numeric questions reward recall more than reasoning: a bond length, the residues per helical turn, a pKₐ, the HAI titer that counts as protective. The table below collects the values tests ask for, grouped by topic. Where textbooks disagree, a range is given; quote the range, or the value your own source uses, and say "about".

ReferenceNumbers, constants and typical values· 177 rows

Lengths in Å (1 Å = 0.1 nm = 10⁻¹⁰ m); energies in kJ/mol with kcal/mol in parentheses (1 kcal = 4.184 kJ). pKₐ and pI values are Lehninger's free-amino-acid values unless noted.

Filter by unit ("Å", "kDa", "°") to see one kind of number at a time.

177 rows

Quantity	Value	Note
Ångström	1 Å = 0.1 nm = 10⁻¹⁰ m	Atomic-scale unit used for all structure distances
Dalton	1 Da = 1 g/mol = 1.66 × 10⁻²⁴ g	1 kDa = 1000 Da
Gas constant R	8.314 J mol⁻¹ K⁻¹ = 1.987 cal mol⁻¹ K⁻¹	Thermodynamics problems
RT at 25 °C (298 K)	2.48 kJ/mol (0.59 kcal/mol)	Thermal energy scale
2.303 RT at 298 K	5.7 kJ/mol (1.36 kcal/mol)	Each 5.7 kJ/mol of ΔG = a factor of 10 in K
Water concentration	55.5 M	Hydrophobic-effect and binding calculations
Dielectric constant	Water ≈ 78–80; protein interior ≈ 2–20 (often 4); hexane ≈ 2	Buried charges interact far more strongly
C–C single bond	1.53–1.54 Å	Freely rotating
C=C double bond	1.34 Å	No rotation
C–H / N–H / O–H	1.09 / 1.01 / 0.96 Å	Hydrogens are not seen in most X-ray structures
C=O (peptide carbonyl)	1.23–1.24 Å	Slightly long because of resonance
Peptide C–N	1.32–1.33 Å	About 40% double-bond character
Amine C–N single bond	1.47–1.49 Å	Compare with the shorter peptide bond
N–Cα / Cα–C (backbone)	1.46 / 1.52 Å	The φ and ψ bonds
C–S bond	1.81–1.82 Å	Cys and Met side chains
S–S (disulfide) bond	2.03–2.05 Å	Longest common covalent bond in proteins
Disulfide Cα–Cα distance	About 4.4–6.8 Å (typically 5.5–6)	Cβ–Cβ about 3.4–4.3 Å; used to test whether two Cys can link
Disulfide C–S–S angle and χ3	About 104°; χ3 ≈ ±90°	Disulfide geometry
Tetrahedral (sp³) angle	109.5°	Cα geometry
Trigonal (sp²) angle	About 120°	Peptide C and N
N–Cα–C angle (τ)	About 111°	Backbone geometry
Cα(i)–Cα(i+1), trans peptide	3.8 Å	Basis of the 2 cm per residue scale
Cα(i)–Cα(i+1), cis peptide	About 2.9 Å	Cis peptides are rare (mostly before Pro)
Peptide-bond rotation barrier	About 80–90 kJ/mol (about 20 kcal/mol)	Why ω is fixed
Trans : cis ratio	About 1000 : 1 for non-Pro peptide bonds; about 4 : 1 in short unfolded peptides for X–Pro, but only about 5–6% cis in folded proteins	Cis-Pro in turns
Van der Waals radii (Bondi)	H 1.20, C 1.70, N 1.55, O 1.52, S 1.80 Å	Spacefill sphere sizes
Van der Waals contact distance	About 3.3–4.0 Å between heavy atoms	Closer = clash
Water-probe radius	1.4 Å	Solvent-accessible surface calculations
Hydrogen bond, donor to acceptor (N···O, O···O)	2.7–3.2 Å (cutoff about 3.5 Å)	Measure in Jmol to confirm an H-bond
Hydrogen bond, H to acceptor	About 1.8–2.2 Å	Near-linear N–H···O is strongest
Salt bridge distance	Charged N to O within about 4 Å	Standard cutoff for calling a salt bridge
Aromatic stacking distance	About 3.4–3.8 Å between ring planes	π-stacking
Zn–S (Cys) / Zn–N (His) in zinc fingers	About 2.3 Å / about 2.0–2.1 Å	Tetrahedral Zn²⁺ site
Covalent C–C bond energy	About 350 kJ/mol (83 kcal/mol)	Covalent ≫ noncovalent
Disulfide S–S bond energy	About 250 kJ/mol (60 kcal/mol)	Covalent cross-link, far stronger than any noncovalent interaction
Hydrogen bond energy	About 4–20 kJ/mol (1–5 kcal/mol) in isolation; net only about 2–8 kJ/mol (0.5–2 kcal/mol) in water	Water competes for H-bonds
Ionic interaction (salt bridge)	Large in vacuum; net in water usually only about 0–12 kJ/mol (0–3 kcal/mol), and a buried salt bridge can even be net destabilizing	Desolvation cancels much of it
Van der Waals interaction	About 0.4–4 kJ/mol (0.1–1 kcal/mol) per atom pair	Weak each, large in total
Hydrophobic burial	About 0.1 kJ/mol per Ų buried (about 25 cal/mol/Ų)	Rough rule for the hydrophobic effect
α helix: residues per turn	3.6	100° per residue
α helix: rise per residue / pitch	1.5 Å / 5.4 Å	Pitch = 3.6 × 1.5 Å
α helix: H-bond and ring	C=O(i) → N–H(i + 4); 13-atom ring (3.6₁₃)	Every backbone N–H and C=O inside the helix is used
α helix: φ, ψ	About −57°, −47° (often −60°, −45°)	Lower-left Ramachandran region
α helix: Cα radius / diameter with side chains	About 2.3 Å / about 10–12 Å	Interior filled by backbone atoms in contact (no hole)
Typical α-helix length	About 10–12 residues (about 3 turns, about 15–18 Å); range 4 to more than 40	Globular proteins
Helix dipole	About +0.5 charge unit at the N-terminus, −0.5 at the C-terminus (effective)	Capping preferences
3₁₀ helix	3.0 residues/turn, 2.0 Å rise, 6.0 Å pitch, i → i + 3, φ/ψ ≈ −49°/−26°	Usually 1–2 turns
π helix	About 4.1–4.4 residues/turn, about 1.15 Å rise, about 5 Å pitch, i → i + 5, φ/ψ ≈ −57°/−70°	Usually a π-bulge
Polyproline II helix	3.0 residues/turn, 3.1 Å rise, 9.3 Å pitch, left-handed, φ/ψ ≈ −75°/+145°	No internal H-bonds
Collagen chain	About 3.3 residues/turn, about 2.9 Å rise per residue; three chains in a right-handed superhelix	Gly every 3rd residue
Collagen molecule	About 300 nm long, 1.5 nm wide; about 1000 residues per chain	Fibrous protein
Coiled coil	3.5 residues per turn of the helix relative to the supercoil (heptad = 2 turns)	Leu at every 7th residue in leucine zippers
β strand: rise per residue	About 3.2–3.5 Å (antiparallel ≈ 3.4–3.5, parallel ≈ 3.2)	Nearly fully extended (max ≈ 3.6–3.8 Å)
β strand: two-residue repeat	Antiparallel ≈ 7.0 Å; parallel ≈ 6.5 Å	Side chains alternate faces
β strand: φ, ψ	Antiparallel ≈ −139°, +135°; parallel ≈ −119°, +113°	Upper-left Ramachandran region
Distance between adjacent strands	About 4.8 Å (4.7–5.0)	Also the 4.7 Å amyloid reflection
Distance between stacked sheets	About 10 Å	β sandwiches; amyloid 10 Å reflection
Typical strand length	About 5–10 residues	Globular proteins
β turn	4 residues, C=O(i) → N–H(i + 3), Cα(i)–Cα(i + 3) < 7 Å	Type II: Gly at i + 2
Type I β turn φ/ψ	i + 1: −60°, −30°; i + 2: −90°, 0°	Most common turn type
Type II β turn φ/ψ	i + 1: −60°, +120°; i + 2: +80°, 0°	i + 2 positive φ, usually Gly
γ turn	3 residues, i → i + 2 H-bond	Rare
Proline φ	About −63° ± 15°	Ring locks φ
Left-handed helix (αL) φ, ψ	About +57°, +47°	Mostly Gly
Secondary-structure content of globular proteins	Roughly 30–35% helix, 20–25% strand, the rest turns and coil	Averages over the PDB
Transmembrane helix	About 20 residues (18–25), spanning about 30 Å of hydrocarbon core	Hydropathy plot window about 19 residues
Side-chain pKₐ (Lehninger, free amino acid)	Asp 3.65, Glu 4.25, His 6.00, Cys 8.18, Tyr 10.07, Lys 10.53, Arg 12.48	Seven ionizable side chains
Side-chain pKₐ (typical in proteins, Berg)	Asp 4.1, Glu 4.1, His 6.0, Cys 8.3, Tyr 10.9, Lys 10.8, Arg 12.5	Textbooks differ by a few tenths
α-COOH / α-NH₃⁺ pKₐ (free amino acids)	About 1.8–2.4 / about 8.8–11.0 (Pro 10.96)	Every amino acid
C-terminus / N-terminus pKₐ in peptides	About 3.1–3.7 / about 7.7–8.0	Lower than in free amino acids
pKₐ shifts inside proteins	Often 1–3 units, occasionally more	Buried or paired groups
pI of amino acids	Gly 5.97; Asp 2.77 (lowest); Glu 3.22; His 7.59; Lys 9.74; Arg 10.76 (highest)	Most neutral residues fall about 5.0–6.3 (Cys 5.07 lowest, Pro 6.30 highest)
pI of example proteins	Pepsin ≈ 1; serum albumin ≈ 4.7–4.9; hemoglobin ≈ 6.8–7.1; cytochrome c ≈ 10; lysozyme ≈ 11	Acidic vs basic proteins
Charge at pH = pKₐ ± 1	About 91% / 9% deprotonated	Henderson–Hasselbalch rule of thumb
Charge at pH = pKₐ ± 2	About 99% / 1% deprotonated	Treat as fully (de)protonated
Physiological pH	Blood 7.35–7.45; cytosol about 7.2	Charges "at pH 7"
Endosome / lysosome pH	Early endosome about 6–6.5, late endosome about 5–5.5; lysosome about 4.5–5	HA fusion is triggered in the endosome
HA fusion pH	About 5.0–6.0, depending on strain	Lower-stability HAs fuse at higher pH
Average residue mass	About 110 Da (abundance-weighted)	Residues ≈ mass ÷ 110
Water lost per peptide bond	18.02 Da	Mass of chain = sum of free amino acids − 18.02 × (n − 1)
Smallest / largest residue	Gly 57.05 Da / Trp 186.21 Da	Mass extremes
Protein density / partial specific volume	About 1.35 g/cm³ / about 0.73 cm³/g	About 1.2 ų per Da
Radius of a compact protein	$R_{min} \approx 0.066\,M^{1/3}$ nm (M in Da): 50 kDa ≈ 2.4 nm	Sphere estimate
Typical domain size	About 50–250 residues, most often about 100–150	Domain definition questions
Typical protein length	Median human protein about 375–415 residues (about 40–45 kDa)	"Normal-sized protein"
Insulin	51 residues (A chain 21, B chain 30), 5.8 kDa, 3 disulfides	First protein sequenced (Sanger)
Ubiquitin	76 residues, 8.6 kDa	Small, very stable
Cytochrome c (human/horse)	104 residues, about 12 kDa	Mass ÷ 110 ≈ 109 residues
Hen egg-white lysozyme	129 residues, 14.3 kDa, 4 disulfides	Classic X-ray structure
Ribonuclease A	124 residues, 4 disulfides; 105 ways to pair 8 Cys	Anfinsen's refolding experiment (random pairing ≈ 1% active)
Myoglobin	153 residues (sperm whale), about 17 kDa, 1 heme	First protein crystal structure (Kendrew, 1958)
Hemoglobin	About 64.5 kDa; α chain 141, β chain 146 residues; 4 hemes	Sickle cell: β Glu6 → Val
GFP	238 residues, about 27 kDa, 11-strand β barrel	β-barrel example
Titin	About 34,000 residues, about 3.8 MDa	Largest known human protein
ΔG of unfolding (net stability)	About 20–60 kJ/mol (5–15 kcal/mol)	Marginal stability
Folding time	Microseconds to seconds (fastest small domains ≈ 1 μs)	Kinetics questions
Levinthal estimate	100 residues × about 3 states per residue gives about 3¹⁰⁰ ≈ 5 × 10⁴⁷ conformations; Levinthal's own figure for 150 residues was about 10³⁰⁰	Random search would take longer than the age of the universe
ΔC_p of unfolding	About 50–60 J mol⁻¹ K⁻¹ per residue (12–14 cal)	Larger for bigger proteins
Typical T_m	Most mesophilic proteins about 45–75 °C; thermophile proteins often > 80 °C	Body temperature 37 °C
Urea / guanidinium chloride for full denaturation	About 6–8 M / about 6 M	Midpoints usually 2–5 M
SDS binding	About 1.4 g SDS per g protein (about 1 SDS per 2 residues)	Uniform charge-to-mass ratio
First protein structures	Myoglobin 1958 (Kendrew), hemoglobin 1959–60 (Perutz); Nobel 1962	History questions
Other milestone dates	α helix and β sheet proposed 1951 (Pauling, Corey); Ramachandran plot 1963; Hodgkin Nobel 1964; Anfinsen Nobel 1972; PDB founded 1971; AlphaFold2 Nobel 2024	History questions
PDB size	About 250,000–260,000 experimental entries (2026), about 80% X-ray overall	Grows by about 15,000–17,000 per year; cryo-EM is now about 40% of new entries
AlphaFold Database	More than 200 million predicted structures	Prediction questions
Resolution bands	> 4 Å: domain shapes, helices as rods; 3–3.5 Å: chain traceable, bulky side chains; about 2.5 Å: most side chains placed; about 2 Å: waters, side-chain orientation; < 1.2 Å: individual atoms, some H	Smaller number = more detail
3UBE resolution	2.15 Å	Good-quality X-ray structure
R-work	About 0.15–0.25 for good structures (rule of thumb R ≈ resolution ÷ 10)	A random model gives about 0.59
R-free	Usually 0.02–0.07 above R-work; 5–10% of reflections set aside	Large gap suggests overfitting
B-factor	Core about 10–30 Ų; loops and surface often > 50–60 Ų	B = 20 Ų ↔ rms displacement ≈ 0.5 Å; 80 Ų ↔ ≈ 1 Å
Matthews coefficient	About 1.6–3.5 ų/Da, typically about 2–3 (mean about 2.5)	Solvent fraction ≈ 1 − 1.23/V_M
Crystal solvent content	About 30–70%, typically about 50%	Proteins in crystals are hydrated, nearly native
X-ray wavelength	Cu Kα 1.54 Å (home source); synchrotron often about 0.9–1.0 Å	E (keV) ≈ 12.4 ÷ λ (Å)
Crystal temperature during data collection	About 100 K	Reduces radiation damage
NMR practical size limit	Up to about 30–50 kDa routinely	Larger with special labeling
Cryo-EM	Routinely 2–4 Å; best below 1.5 Å; practical lower size limit about 40–50 kDa; 10⁴–10⁶ particle images	Large complexes
pLDDT bands	> 90 very high; 70–90 confident; 50–70 low; < 50 very low (often disordered)	AlphaFold confidence
RMSD between structures	< 1 Å nearly identical; 1–2 Å same fold, local differences; > 3–4 Å substantially different	Usually on Cα atoms
Homology modeling identity	Reliable above about 30–40%; "twilight zone" about 20–35%	Template choice
Contact-map cutoff	Often 8 Å between Cα (or Cβ) atoms	Contact maps
NOE distance limit	About 5–6 Å	NMR restraints
CD signatures	α helix: minima 208 and 222 nm, maximum about 192 nm; β sheet: minimum about 216–218 nm, maximum about 195 nm; random coil: minimum about 198–200 nm	Far-UV CD (190–250 nm)
Molar absorptivity at 280 nm	Trp 5500, Tyr 1490, cystine 125 M⁻¹ cm⁻¹	Protein concentration from A₂₈₀
Peptide-bond absorbance	Strong below 230 nm (often read at 205–214 nm)	Detecting peptides without Trp/Tyr
Trp fluorescence	Excite at 280 or 295 nm; emission about 330 nm buried, about 350 nm exposed	Red shift on unfolding
His-tag	Usually 6 His (6–10); Ni²⁺–NTA resin; eluted with about 100–500 mM imidazole	Purification
Edman degradation	One residue per cycle; practical length about 30–50 residues	Sequence peptides, not whole proteins
Genetic code	64 codons = 61 sense + 3 stop (UAA, UAG, UGA); 1 start (AUG)	Codon-counting questions
Codons per amino acid	6: Leu, Ser, Arg; 4: Gly, Ala, Val, Pro, Thr; 3: Ile; 2: Phe, Tyr, His, Gln, Asn, Lys, Asp, Glu, Cys; 1: Met, Trp	Degeneracy
Amino acid counts	20 standard; 22 genetically encoded (+ Sec at UGA, Pyl at UAG); 9 essential in humans	"How many" questions
Residues with an ionizable side chain	7 (D, E, H, C, Y, K, R) → three pKₐ values; the other 13 have two	Titration questions
Possible sequences	20ⁿ for n residues (a 100-mer: 20¹⁰⁰ ≈ 10¹³⁰)	Combinatorics questions
Peptide bonds in a chain	n − 1 for n residues	Water released = n − 1
Human protein-coding genes	About 19,000–20,000	Proteome size
Translation speed	About 15–20 residues/s in bacteria; about 5–6 residues/s in eukaryotes	Cotranslational folding
Phosphoproteome	About one third of human proteins are phosphorylated; about 500 human kinases	PTM prevalence
N-terminal acetylation	About 80–85% of human proteins	Most common PTM by protein count
Model scale	2 cm per residue ÷ 3.8 Å ≈ 0.53 cm per Å (magnification ≈ 5 × 10⁷)	Converting any real distance to model cm
3UBE model segment	Chain A residues 153–252 = 100 residues = 200 cm of backbone	Cut a little over 2 m of material
Model helix	About 0.8 cm rise per residue; about 2.8 cm per turn; about 2.4 cm Cα-cylinder diameter	1.5, 5.4 and 4.6 Å scaled
Model strand	About 1.7–1.8 cm advance per residue	3.3–3.4 Å scaled; nearly straight
Model specifications	Backbone diameter 0.75–1.25 cm; fits in 50.0 × 50.0 × 50.0 cm; legend 8.5 × 5.5 in	2027 rules
Influenza A virion	About 80–120 nm when spherical; filaments can exceed several μm	Pleomorphic
Influenza genome	8 segments of (−)ssRNA, about 13.5 kb in total (13,588 nt in A/PR/8/34)	Segment 1–2: 2341 nt; segment 8: 890 nt
Influenza A proteins	10 core proteins from 8 segments, plus accessory proteins (PB1-F2, PA-X and others), up to about 17	Segments 7 and 8 each give two proteins by splicing
Spikes per virion	HA about 300–500 trimers; NA about 40–100 tetramers; M2 only about 15–70 molecules (a few tetramers)	HA:NA roughly 4–10 : 1
HA of A/California/04/2009	566-residue precursor: 17-residue signal peptide, HA1 = 327 residues, HA2 = 222 residues (H1 numbering)	Polypeptide about 63 kDa, more with glycans
HA trimer size	About 135 Å long from membrane to head	Spike length about 13–14 nm
HA fusion-peptide length	About 20 residues at the HA2 N-terminus	Exposed by HA0 cleavage
HA/NA subtypes	H1–H18 and N1–N11 (H17N10 and H18N11 only in bats)	Subtype naming; a candidate H19 from wild ducks was reported in 2023
Influenza RdRp error rate	About 1 per 10⁴–10⁵ nucleotides copied	About one mutation per genome copy
Reassortant genotypes from 2 parents	2⁸ = 256 (254 new)	Each segment is an independent choice
Sialic acid	9 carbons (Neu5Ac)	Receptor sugar
H1 antigenic sites	5 (Sa, Sb, Ca1, Ca2, Cb); H3 also 5 (A–E)	Mostly on HA1
Incubation period	1–4 days, average about 2	CDC
Infectious period	From about 1 day before symptoms to 5–7 days after (longer in children)	CDC
Antiviral treatment window	Best within 48 h of symptom onset	Neuraminidase inhibitors
Vaccine response time	About 2 weeks after vaccination for protective antibodies	Vaccine timing
Vaccine HA dose	15 μg HA per strain (standard); 60 μg (high-dose, 65+); 45 μg (recombinant)	Inactivated vaccines
Global seasonal burden (WHO)	About 1 billion cases, 3–5 million severe cases, 290,000–650,000 respiratory deaths per year	Public-health context
Pandemic years	1918 H1N1, 1957 H2N2, 1968 H3N2, 2009 H1N1pdm09	Antigenic shift
1918 deaths	About 50 million worldwide (about 675,000 US)	Deadliest pandemic
2009 pandemic deaths, first year	About 151,700–575,400 worldwide (CDC model)	Mild compared with 1918
IgG	About 150 kDa: 2 heavy (about 50 kDa) + 2 light (about 25 kDa) chains; 12 Ig domains; 2 binding sites	Fab and Fc each about 50 kDa
Ig domain	About 100–110 residues, one conserved disulfide	Ig fold
CDRs	6 per binding site (3 heavy, 3 light)	CDR-H3 most variable
IgM	Pentamer, about 970 kDa, 10 binding sites, with J chain	First antibody in a primary response
Antibody classes / human IgG subclasses	5 classes (IgM, IgG, IgA, IgE, IgD) / 4 IgG subclasses	Isotypes
IgG share and half-life	About 75–80% of serum antibody; half-life about 3 weeks	Passive immunity duration
Antibody affinity	K_d typically 10⁻⁷–10⁻¹¹ M after affinity maturation	Lower K_d = tighter
Antibody–antigen footprint	About 15–22 residues per side, about 600–900 Ų buried	Conformational epitopes
MHC peptide length	Class I about 8–10 residues; class II about 13–25	Closed vs open groove
Primary vs secondary response	Primary: detectable antibody after about 7–10 days, IgM first, IgG peaking in 2–4 weeks; secondary: faster (days) and higher, mainly IgG	Memory
Protective HAI titer	≥ 1:40, about 50% protection	Seroprotection
Seroconversion	≥ 4-fold rise in HAI titer to at least 1:40 (or < 1:10 before to ≥ 1:40 after)	Vaccine trials
Complement	About 30 proteins, 3 activation pathways	Innate immunity
Interferon types	3 (I: α/β; II: γ; III: λ)	Antiviral signaling

When a test question gives its own constants (a pKₐ, a ΔC_p, a wavelength), use the question's values, not these. These are the defaults to fall back on when it does not.

Named proteins tests reuse

Test writers draw on a small, recurring set of proteins: the ones that appear in every biochemistry textbook because each one was the first, the clearest or the most medically important example of some principle. Myoglobin was the first structure, insulin the first sequence, ribonuclease A the refolding experiment, lysozyme the first enzyme structure, chymotrypsin the catalytic triad, hemoglobin cooperativity, collagen the triple helix. When a question names one of these proteins, it is usually not testing that protein's details. It is using the protein as a handle for the principle it is famous for. Learn the "why tests use it" column as carefully as the "what it does" column.

When a question names a protein you do not recognize, work from its class. The name usually gives the class away: "-ase" is an enzyme, and the root names the substrate (lactase splits lactose, protease cuts proteins, kinase adds phosphate, phosphatase removes it, polymerase builds polymers). "-in" proteins are often structural, transport or signaling proteins (keratin, globin, tubulin). "Receptor", "channel", "pump" or "transporter" means a membrane protein with hydrophobic transmembrane helices. Fibrous proteins (long repeats, high Gly/Pro/Ala) are structural and insoluble; globular proteins are soluble and do the chemistry. From the class you can usually predict the fold, the location, what a mutation would break and which technique would study it. Rows marked context only describe proteins whose internal details fall outside the 2027 rules; know the one-line identity, not the mechanism.

ReferenceNamed proteins and what they illustrate· 123 rows

Rows are grouped roughly by theme: oxygen carriers and fibrous proteins, motors, classic enzymes, membrane proteins, blood and immune proteins, folding and misfolding, regulation and the genome, viral proteins, and influenza. "Context only" means the protein's internal details are outside the 2027 rules.

123 rows

Protein	Class / fold	What it does	Why tests use it
Hemoglobin	Globular, all-α globin fold; α₂β₂ tetramer, 4 hemes	Carries O₂ in red blood cells from lungs to tissues	Quaternary structure, cooperativity (sigmoidal curve), T/R states, Bohr effect, 2,3-BPG; sickle cell (β Glu6→Val)
Myoglobin	Globular, all-α globin fold (8 helices, A–H), one heme	Stores O₂ in muscle	First protein structure (Kendrew, 1958); hyperbolic binding curve; contrast with hemoglobin
Fetal hemoglobin (HbF)	α₂γ₂ tetramer	Carries O₂ in the fetus with higher affinity than adult Hb	Binds 2,3-BPG weakly, so it pulls O₂ from maternal blood
Collagen	Fibrous; three left-handed polyproline-II-like strands wound into a right-handed triple helix	Tensile strength in skin, tendon, bone, cartilage	Gly-X-Y repeat (Gly every third residue fits the crowded center), hydroxyproline needs vitamin C (scurvy), most abundant mammalian protein
α-Keratin	Fibrous; α-helical coiled coil (two right-handed helices in a left-handed supercoil)	Hair, nails, wool, outer skin	Coiled coil; many disulfide cross-links (hair perms break and re-form them)
β-Keratin	Fibrous; β-sheet	Feathers, scales, claws of birds and reptiles	Contrast with α-keratin: same use, different secondary structure
Elastin	Fibrous, largely disordered chains cross-linked through modified lysines (desmosine)	Stretch and recoil in arteries, lungs, skin	Elasticity is entropic: stretching orders the chains, release restores disorder
Silk fibroin	Fibrous; stacked antiparallel β-sheets	Silk of silkworms and spiders	Gly-Ala-Gly-Ala-Gly-Ser repeat: small side chains let sheets pack tightly; strong but not stretchy
Actin	Globular G-actin polymerizes to F-actin filaments (uses ATP)	Microfilaments: cell shape, movement, muscle thin filaments	Polymerization of a globular protein into a filament; partner of myosin
Myosin	Motor protein; globular heads plus long coiled-coil tail	Walks along actin using ATP; muscle contraction	Chemical energy (ATP hydrolysis) converted to mechanical work; coiled-coil tail
Tubulin	Globular α/β heterodimer (binds GTP)	Builds microtubules: spindle, cilia, intracellular tracks	Dynamic instability; target of colchicine and taxol
Kinesin and dynein	Motor proteins	Carry cargo along microtubules (kinesin mostly toward the plus end, dynein toward the minus end)	Motor proteins other than myosin; ATP-driven stepping
Titin	Giant chain of Ig-like and fibronectin-like domains	Molecular spring that positions myosin in the sarcomere	Largest known protein (about 34,000 residues, roughly 3–4 MDa)
Spectrin	Long α-helical repeats	Mesh under the red-cell membrane that keeps cells flexible	Membrane skeleton; mutations give misshapen red cells
Tropomyosin and troponin	Coiled coil (tropomyosin); Ca²⁺-binding (troponin)	Block and unblock myosin sites on actin in response to Ca²⁺	Calcium switch for muscle contraction
Insulin	Small globular hormone; A chain (21 aa) and B chain (30 aa) joined by 2 interchain disulfides plus 1 intrachain disulfide	Lowers blood glucose by signaling cells to take it up	First protein sequenced (Sanger); made as proinsulin, C-peptide removed; first recombinant human drug (1982)
Proinsulin	Single chain: B chain, C-peptide, A chain	Precursor folded in the ER	Shows that processing after folding can leave a two-chain protein that will not refold well on its own
Glucagon	29-residue peptide hormone	Raises blood glucose	Opposite of insulin; peptide hormone example
Oxytocin and vasopressin	9-residue cyclic peptides (one disulfide)	Hormones (uterine contraction; water retention)	Very small peptides with a disulfide ring; first peptide hormones synthesized (du Vigneaud)
Ribonuclease A (RNase A)	Globular α+β, 124 aa, 4 disulfides	Cleaves RNA	Anfinsen's refolding experiment: sequence determines structure; His12 and His119 act as acid and base
Lysozyme (hen egg white)	Globular α+β, 129 aa, 4 disulfides	Cleaves the bond between NAM and NAG sugars in bacterial cell walls	First enzyme structure (Phillips, 1965); Glu35 and Asp52 catalysis; a standard crystal in teaching labs
Chymotrypsin	Serine protease; two β-barrel domains	Cuts after large hydrophobic residues (Phe, Tyr, Trp)	Catalytic triad Ser195–His57–Asp102, oxyanion hole, zymogen chymotrypsinogen
Trypsin	Serine protease (chymotrypsin fold)	Cuts after Lys and Arg	Asp189 at the bottom of the specificity pocket pairs with the + side chain; activates other pancreatic zymogens
Elastase	Serine protease (chymotrypsin fold)	Cuts after small residues (Ala, Gly, Val)	Shallow pocket blocked by Val and Thr: same fold, different specificity
Subtilisin	Bacterial serine protease, different fold	Digests proteins; used in detergents	Same Ser–His–Asp triad in an unrelated fold: convergent evolution
Thrombin	Serine protease of blood clotting	Cuts fibrinogen to fibrin; activates platelets	Protease cascade; zymogen prothrombin; anticoagulant target
Fibrinogen and fibrin	Fibrous, long coiled-coil arms	Fibrinogen is soluble; once cut, fibrin polymerizes into the clot mesh	Proteolytic activation turns a soluble protein into a polymer
Plasmin	Serine protease	Dissolves fibrin clots	Clot breakdown; activated by tPA (a clot-busting drug)
Pepsin	Aspartic protease (two Asp in the active site)	Digests proteins in the stomach	Works best near pH 2; pepsinogen self-activates in acid; contrast with serine proteases
Papain and caspases	Cysteine proteases (Cys–His dyad)	Papain digests proteins (meat tenderizer); caspases execute apoptosis	Cysteine thiol as the nucleophile instead of serine
Carboxypeptidase A	Zinc metalloprotease	Removes the C-terminal residue of a peptide	Zn²⁺ activates water; exopeptidase versus endopeptidase
Enteropeptidase (enterokinase)	Serine protease of the gut wall	Cuts trypsinogen to trypsin	Starts the zymogen activation cascade
α₁-Antitrypsin (a serpin)	Metastable serpin fold	Traps and inhibits elastase in the lungs	Metastable native state that springs into a more stable form; deficiency causes emphysema
Bovine pancreatic trypsin inhibitor (BPTI)	Small α+β, 58 aa, 3 disulfides	Blocks trypsin	Classic folding model (disulfide intermediates) and early NMR and simulation subject
Lipocalins	8-stranded β-barrel with a cup for small molecules	Carry hydrophobic ligands; some act as protease inhibitors	β-barrel binding pocket
Carbonic anhydrase	Zinc enzyme, mostly β-sheet	CO₂ + H₂O ⇌ HCO₃⁻ + H⁺	Among the fastest enzymes (about 10⁶ reactions per second); Zn²⁺ bound by three His activates water
Catalase	Heme enzyme, tetramer	2 H₂O₂ → 2 H₂O + O₂	One of the highest turnover numbers known (millions per second); the fizzing peroxide demo
Superoxide dismutase (SOD1)	Cu/Zn enzyme, β-barrel dimer	Converts superoxide to H₂O₂ and O₂	Antioxidant defense; misfolding mutants cause familial ALS
Urease	Nickel enzyme	Urea → ammonia + CO₂	First enzyme crystallized (Sumner, 1926): proved enzymes are proteins
Hexokinase	Two-lobed α/β enzyme	Phosphorylates glucose	Induced fit: the lobes clamp shut around glucose
Triose phosphate isomerase (TIM)	TIM barrel (8 α/β units)	Interconverts glycolysis three-carbon sugars	Named the TIM-barrel fold; a "catalytically perfect" (diffusion-limited) enzyme
Lactate dehydrogenase	Rossmann fold (NAD⁺-binding), tetramer	Pyruvate ⇌ lactate	Rossmann fold for nucleotide binding; isozymes
Aspartate transcarbamoylase (ATCase)	Allosteric enzyme with catalytic and regulatory subunits	First step of pyrimidine synthesis	Feedback inhibition by CTP; allostery and T/R states in an enzyme
Acetylcholinesterase	α/β hydrolase, serine triad	Breaks down acetylcholine at synapses	Very fast enzyme; irreversibly inhibited by nerve agents and organophosphates
Dihydrofolate reductase (DHFR)	α/β enzyme	Makes tetrahydrofolate for DNA building blocks	Competitive inhibition by methotrexate (drug design example)
Rubisco	Large multi-subunit enzyme	Fixes CO₂ in photosynthesis	Most abundant protein on Earth; slow and confuses O₂ for CO₂
Lactase	β-galactosidase enzyme	Lactose → glucose + galactose	Lactose intolerance versus lactase persistence; specificity from the substrate name
Amylase	Glycoside hydrolase (TIM barrel)	Cuts α-1,4 links in starch (saliva and pancreas)	Digestive enzyme specific to one bond type; cannot cut cellulose's β-1,4 links
Lipase	α/β hydrolase, serine triad	Breaks fats (triglycerides) into fatty acids and glycerol	Works at an oil–water interface; serine triad outside the proteases
Sucrase-isomaltase	Brush-border glycosidase	Splits sucrose and maltose	Digestive enzymes named after substrates
Firefly luciferase	Enzyme (adenylate-forming)	Luciferin + ATP + O₂ → light	Reporter gene and ATP assay; bioluminescence is enzyme-made light
Green fluorescent protein (GFP)	11-stranded β-barrel around a central helix	Fluoresces green under blue light	Chromophore forms by itself from Ser65–Tyr66–Gly67; fusion tag for tracking proteins (Nobel 2008)
Streptavidin	β-barrel tetramer	Binds biotin	One of the tightest non-covalent bonds known; used in labeling and detection
Rhodopsin	GPCR, 7 transmembrane helices	Absorbs light in rod cells and triggers a G-protein cascade	Retinal (11-cis → all-trans) bound to a Lys by a Schiff base; light becomes a nerve signal
Bacteriorhodopsin	7 transmembrane helices (not a GPCR)	Light-driven proton pump in archaea	First membrane protein structure, by electron crystallography (Henderson and Unwin, 1975)
β₂-Adrenergic receptor	GPCR	Responds to adrenaline	First structure of a GPCR bound to its G protein (Kobilka, 2011); about a third of drugs target GPCRs
Heterotrimeric G protein (Gα, Gβ, Gγ)	GTPase switch	Relays a signal from a GPCR	GTP = on, GDP = off; hydrolysis turns itself off
Ras	Small GTPase	Relays growth signals	Molecular switch; oncogenic mutations (such as at Gly12) lock it "on"
Insulin receptor and EGFR	Receptor tyrosine kinases (one transmembrane helix)	Hormone binding activates the inside kinase	Dimerization and autophosphorylation; contrast with GPCRs
K⁺ channel (KcsA)	Tetrameric ion channel	Conducts K⁺ at least a thousand times more selectively than the smaller Na⁺	Selectivity filter (TVGYG) lined with backbone carbonyls (MacKinnon, Nobel 2003)
Voltage-gated Na⁺ channel	Ion channel with voltage sensors	Fires the action potential	Gating by membrane voltage; blocked by tetrodotoxin and local anesthetics
Nicotinic acetylcholine receptor	Ligand-gated ion channel (pentamer)	Opens on acetylcholine binding at nerve–muscle junctions	Ligand-gated versus voltage-gated channels
Aquaporin	Tetramer, 6 transmembrane helices each	Water channel; excludes H⁺	Two NPA motifs in the pore; Agre, Nobel 2003
Na⁺/K⁺-ATPase	P-type ATPase pump	Pumps 3 Na⁺ out and 2 K⁺ in per ATP	Primary active transport; Skou, Nobel 1997
GLUT1	Major facilitator transporter, 12 transmembrane helices	Moves glucose down its gradient	Facilitated diffusion (uniporter); saturates, unlike simple diffusion
CFTR	ABC transporter family, but works as a Cl⁻ channel	Moves chloride across epithelial membranes	ΔF508 misfolds and is degraded in the ER (cystic fibrosis)
ATP synthase	Rotary F₀F₁ complex in membranes	Uses the proton gradient to make ATP	Rotary motor; binding-change mechanism (Boyer and Walker, Nobel 1997)
Cytochrome c	Small heme protein	Carries electrons in the respiratory chain; triggers apoptosis	Highly conserved: classic for sequence comparison and evolutionary trees
Antibody (IgG)	Ig fold (β-sandwich domains); 2 heavy + 2 light chains, about 150 kDa, Y shape	Binds antigens; recruits immune effectors	Fab (binding) versus Fc (effector); 6 CDR loops; disulfide-linked chains
IgM, IgA, IgE	Antibody classes (pentamer, dimer, monomer)	First response; mucosal; allergy and parasites	Class switching; secretory IgA in the airway matters for influenza
MHC class I and II	Peptide-binding groove of two helices on a β-sheet floor	Present peptides to CD8⁺ (class I) and CD4⁺ (class II) T cells	Antigen presentation; class I shows internal (viral) peptides
T-cell receptor	Ig-fold heterodimer	Recognizes peptide–MHC	Recognizes processed peptides, never free antigen
Interferons (type I)	Small helical cytokines	Put neighboring cells into an antiviral state	First innate response to influenza; NS1 blocks their induction
RIG-I and TLR7	Pattern-recognition receptors	Sense viral RNA (cytosol and endosome)	How innate immunity detects an RNA virus
Mx proteins (MxA)	Dynamin-like GTPase	Interferon-induced block of influenza replication	Example of an interferon-stimulated antiviral protein
Serum albumin	All-α, three domains	Keeps blood osmotic pressure; carries fatty acids, drugs, bilirubin	Most abundant plasma protein (about 66 kDa); nonspecific hydrophobic pockets
Ferritin	24-subunit hollow shell of 4-helix bundles	Stores iron inside cells (thousands of Fe³⁺ per shell)	Quaternary cage; iron storage versus transport
Transferrin	Two-lobed globular glycoprotein	Carries Fe³⁺ in the blood (2 sites)	Receptor-mediated endocytosis; pairs with ferritin
Lectins (for example concanavalin A)	Carbohydrate-binding proteins	Recognize specific sugars	HA is a lectin: binding a glycan, not an enzyme reaction
Mucins	Heavily glycosylated proteins	Form the mucus layer of the airway	Carry sialic acid decoys that trap influenza; NA cuts through them
Erythropoietin (EPO)	Glycoprotein hormone, 4-helix bundle	Stimulates red-cell production	Glycosylation and protein drugs (and doping)
Prion protein (PrP)	Normal PrP^C mostly α-helical; PrP^Sc β-rich amyloid	Normal role unclear; misfolded form is infectious	Templated misfolding; CJD, kuru, BSE, scrapie; Prusiner (Nobel 1997)
Amyloid-β (Aβ)	40–42 residue peptide cut from APP by β- and γ-secretase	Forms extracellular plaques	Alzheimer's disease; cross-β amyloid
Tau	Intrinsically disordered microtubule-associated protein	Stabilizes microtubules in neurons	Hyperphosphorylated tau forms tangles (Alzheimer's)
α-Synuclein	Intrinsically disordered	Synaptic vesicle protein	Lewy bodies in Parkinson's disease
Huntingtin	Large protein with an N-terminal polyglutamine tract	Neuronal functions	Expanded CAG repeat (above about 36) → long polyQ → aggregation (Huntington's)
Amylin (IAPP)	37-residue peptide	Co-secreted with insulin	Pancreatic amyloid in type 2 diabetes
Hsp70 (BiP in the ER)	ATP-driven chaperone	Binds exposed hydrophobic stretches of new chains	Chaperone that binds directly; prevents aggregation
Hsp60 / GroEL–GroES	Chaperonin: two 7-subunit rings plus a lid	Encloses one chain so it folds alone	Chaperonin encloses; chaperone binds (the classic contrast)
Hsp90	ATP-driven chaperone	Late folding of signaling proteins and kinases	Heat-shock protein family; cancer drug target
Protein disulfide isomerase (PDI)	Thioredoxin fold	Forms and reshuffles disulfides in the ER	Folding enzyme that speeds Anfinsen-type refolding in cells
Calnexin and calreticulin	ER lectin chaperones	Hold new glycoproteins (including HA) until folded	Glycan-based quality control
Ubiquitin	Small β-grasp fold, 76 aa	Tagged onto Lys of target proteins by E1–E2–E3 enzymes	Lys48 chains → proteasome degradation; extremely conserved (Nobel 2004)
Proteasome (26S)	Barrel-shaped protease complex	Destroys ubiquitin-tagged proteins	Regulated protein turnover; also makes MHC class I peptides
Calmodulin	Dumbbell of two lobes with EF hands	Binds 4 Ca²⁺ and then grips and activates target proteins	EF-hand motif; calcium sensing; conformational change on binding
Dystrophin	Long spectrin-like rod	Links the muscle cytoskeleton to the membrane	Duchenne muscular dystrophy; encoded by the largest human gene
p53	DNA-binding core domain flanked by disordered regions; tetramer	Tumor suppressor: halts the cell cycle or triggers apoptosis after DNA damage	"Guardian of the genome"; most cancer mutations hit the DNA-binding core (Arg175, Arg248, Arg273)
Histones (H2A, H2B, H3, H4; H1)	Small, Lys/Arg-rich; histone fold	Octamer spools DNA into nucleosomes; H1 links them	Positive tails bind negative DNA; Lys acetylation removes + charge and opens chromatin
Zinc finger transcription factors	ββα fold with Zn²⁺ held by Cys₂His₂	Read DNA bases in the major groove	Metal-stabilized small domain (context only: earlier-season target class)
Leucine zipper (GCN4)	Coiled coil of two helices, Leu every 7th residue	Dimerizes bZIP transcription factors	Heptad repeat and coiled coils
Lac repressor	Helix-turn-helix DNA-binding protein, allosteric	Blocks transcription until lactose is present	Allostery and helix-turn-helix DNA binding
DNA polymerase	Hand-shaped (fingers, palm, thumb)	Copies DNA 5′→3′; many proofread 3′→5′	Enzyme as a protein machine; Taq polymerase is heat-stable (PCR)
RNA polymerase	Multi-subunit enzyme	Transcribes DNA into RNA	Eukaryotic transcription structure (Kornberg, Nobel 2006)
Ribosomal proteins	Many small, mostly basic proteins	Scaffold rRNA in the ribosome	The ribosome is a ribozyme: rRNA, not protein, forms the peptide bond (Nobel 2009)
Protein kinases (PKA, CDKs)	Two-lobed kinase fold binding ATP	Add a phosphate from ATP to Ser, Thr or Tyr	Phosphorylation as an on/off switch; CDK1 with cyclin drives the cell cycle; about 500 human kinases
BCR-ABL kinase	Fusion tyrosine kinase	Constantly active in chronic myeloid leukemia	Imatinib: a rational, structure-based kinase inhibitor
Protein phosphatases	Several folds (PP1, PP2A, PTP1B)	Remove phosphate groups	Reverse of kinases; reversible phosphorylation (Krebs and Fischer, Nobel 1992)
Glycogen phosphorylase	Allosteric dimer	Releases glucose from glycogen	Controlled by both phosphorylation and allostery
SARS-CoV-2 spike	Class I fusion glycoprotein trimer, like HA	Binds ACE2 and fuses membranes	Context only: same fusion logic as HA (cleavage, prefusion → postfusion); D614G as a mutation-naming example
HIV gp120/gp41 (Env)	Class I fusion glycoprotein trimer	Binds CD4 and coreceptor; gp41 fuses membranes	Context only: heavy glycan shield, another HA-like fusion protein
HIV protease	Aspartic protease, homodimer (99 aa per chain)	Cuts the Gag and Gag-Pol polyproteins into working proteins	Each monomer contributes one Asp25; landmark structure-based drug design (protease inhibitors)
HIV reverse transcriptase	p66/p51 heterodimer with polymerase and RNase H domains	Copies RNA genome into DNA	Error-prone, so HIV mutates fast; AZT (nucleoside) and non-nucleoside inhibitors
HCV NS3/4A protease	Serine protease (chymotrypsin-like)	Cuts the HCV polyprotein	Context only: shallow active site defeated by reversible covalent inhibitors (boceprevir, telaprevir)
Furin and TMPRSS2	Host proteases (subtilisin-like; serine protease)	Cleave viral fusion proteins	Furin cuts multibasic HA sites (highly pathogenic H5/H7); TMPRSS2 cuts seasonal HA0 in the airway
Influenza hemagglutinin (HA)	Class I fusion glycoprotein, homotrimer; HA1 head + HA2 stem	Binds sialic acid; fuses viral and endosomal membranes at low pH	The 2027 protein (3UBE): receptor binding site, cleavage of HA0, fusion peptide, antigenic sites
Influenza neuraminidase (NA)	Tetramer; each head a 6-bladed β-propeller	Cuts sialic acid to release new virions and get through mucus	Target of oseltamivir and zanamivir; H/N subtype naming; HA–NA balance
Influenza M1	Matrix protein, helical domains	Lines the inside of the envelope; organizes budding and vRNP export	Most abundant virion protein; the shell under the membrane
Influenza M2	Single-pass membrane protein, tetrameric proton channel	Acidifies the virion interior so vRNPs uncoat	Target of amantadine and rimantadine; S31N resistance
Influenza NP	Nucleoprotein, RNA-binding	Coats every genome segment into an RNP	Conserved T-cell target; defines types A versus B
Influenza NS1	Dimeric RNA-binding and effector domains	Blocks interferon induction (RIG-I) and host mRNA processing	Viral immune evasion protein
Influenza NEP (NS2)	Small nuclear export protein	Exports vRNPs from the nucleus with M1	Why influenza needs nuclear export signals
Influenza PB1, PB2, PA	Heterotrimeric RNA-dependent RNA polymerase	PB1 adds nucleotides; PB2 grabs host 5′ caps; PA's endonuclease cuts them (cap snatching)	Polymerase lacks proofreading (drift); PB2 E627K host adaptation; baloxavir targets PA
Influenza PB1-F2	Short accessory protein from an alternative reading frame	Targets mitochondria, promotes cell death	Overlapping reading frames; truncated in 2009 pandemic viruses

The table is long, but the pattern is short: every row pairs a protein with a principle. If a question names a protein and asks something you do not know, fall back on the principle in the last column, which is usually what the question is about. The influenza rows at the end are the ones that connect directly to 3UBE; for HA itself, see the hemagglutinin chapters rather than this summary.

Techniques and what they tell you

Tests ask about methods in three ways: which technique measures a given property, what a named reagent or step does, and which method fits a given sample (size, purity, whether it crystallizes). The table below is organized by what the method measures, from atomic structure down to simple concentration. Use the middle column to answer "which technique would you use" and the right column for the one detail most likely to be asked.

ReferenceLaboratory and computational techniques· 77 rows

Grouped as: structure determination, sequencing and mass, separation and purification, spectroscopy and concentration, binding and stability, function and immunological assays, and computational methods.

77 rows

Technique	What it measures	Key detail to remember
X-ray crystallography	Electron density of a crystal, hence atomic positions	Needs crystals; phase problem; Bragg's law; quality by resolution, R, R-free; about 80% of the PDB; 3UBE was solved this way
Molecular replacement	Initial phases for a new crystal	Places a similar known structure in the new unit cell (3UBE used 3LZG); works when sequences are similar
Isomorphous replacement (MIR)	Phases from heavy-atom soaks	Heavy atoms (Hg, Pt, U) change spot intensities; used for myoglobin and hemoglobin
Anomalous scattering (SAD/MAD)	Phases from atoms that absorb near the X-ray wavelength	Selenomethionine replaces Met; MAD needs a tunable synchrotron beam, SAD only one wavelength
Fiber diffraction	Repeat distances in aligned fibers	Photo 51 (DNA); α-keratin's 5.1 Å repeat was an early clue to coiled coils
Neutron crystallography	Nuclear positions, including hydrogen and deuterium	Sees H atoms and protonation states that X-rays miss; needs large crystals and a reactor or spallation source
NMR spectroscopy	Chemical shifts and through-space distances of nuclei (¹H, ¹⁵N, ¹³C) in solution	Ensemble of models; dynamics; best for proteins under about 30–50 kDa; isotope labeling
NOE / NOESY	Distances between hydrogens closer than about 5–6 Å	The main source of distance restraints for NMR structures
¹H–¹⁵N HSQC	One peak per backbone amide (prolines have none)	"Fingerprint" of a folded protein; peaks shift when a ligand binds
Hydrogen–deuterium exchange (NMR or MS)	How fast amide hydrogens swap with D₂O	Buried and H-bonded amides exchange slowly; maps structure and stability
Cryo-electron microscopy (single particle)	Density map from many images of frozen particles	Vitreous ice; particle picking, 2D classes, 3D reconstruction; best for large complexes; Nobel 2017
Cryo-electron tomography	3D images of objects in place, such as whole virions	Tilt series of one sample; shows HA and NA spikes on intact flu particles
Electron crystallography and MicroED	Structures from 2D crystals or tiny 3D crystals by electron diffraction	Bacteriorhodopsin (1975, 1990); MicroED works on crystals too small for X-rays
Small-angle X-ray scattering (SAXS)	Overall size and shape in solution	Low resolution (envelope only); radius of gyration; checks flexibility and oligomer state
Negative-stain electron microscopy	Shape of particles stained with heavy metal	Quick, low resolution; classic images of influenza virions
Atomic force microscopy and optical tweezers	Forces needed to unfold single molecules	Single-molecule mechanics, such as titin domains unfolding one at a time
Edman degradation	N-terminal sequence, one residue per cycle	(1) Phenyl isothiocyanate (PITC) couples to the free N-terminal amine at mildly basic pH; (2) anhydrous trifluoroacetic acid cleaves off the first residue as a thiazolinone; (3) it is converted to a stable PTH-amino acid and identified by HPLC. The rest of the chain is untouched; about 30–50 residues per run; fails if the N-terminus is blocked (acetylated)
Sanger N-terminal labeling	Identity of the N-terminal residue	Fluorodinitrobenzene (FDNB, "Sanger's reagent") tags the N-terminus; hydrolysis then destroys the chain, so only one residue per experiment
Dansyl chloride labeling	Identity of the N-terminal residue	Fluorescent tag, more sensitive than FDNB; also destroys the chain
Amino acid analysis	Amino acid composition	Acid hydrolysis (6 M HCl, about 110 °C, 24 h) then chromatography and ninhydrin; destroys Trp and converts Asn/Gln to Asp/Glu
Chemical and enzymatic cleavage	Fragments for sequencing or mapping	Trypsin after Lys/Arg; chymotrypsin after Phe/Tyr/Trp; cyanogen bromide after Met; neither protease cuts when the next residue is Pro
Overlapping-fragment sequencing	Full sequence from two different digests	Overlaps between fragments from two different cleavage methods give the order (Sanger's insulin strategy)
Disulfide reduction and alkylation	Breaks and caps disulfide bonds	β-mercaptoethanol or DTT reduce S–S; iodoacetamide caps Cys so the bonds cannot re-form
Mass spectrometry (ESI, MALDI)	Mass-to-charge ratio of ionized molecules	Exact mass, sequence and modifications; ESI and MALDI ionize intact proteins (Nobel 2002)
Tandem MS (MS/MS)	Sequence from fragmentation of peptides	Breaking peptide bonds gives b- and y-ion ladders that spell the sequence
Peptide mapping and peptide mass fingerprinting	Pattern of masses after digestion	Compared against a database to identify a protein or find a mutation or modification
SDS-PAGE	Apparent molecular mass of denatured chains	SDS unfolds and coats chains with uniform negative charge so they separate by size; small chains run fastest; add reducing agent to break disulfides (Laemmli, 1970)
Native PAGE	Size, shape and charge of folded proteins	No SDS: subunits stay together, so it reports the intact complex; compare with SDS-PAGE to count subunits
Isoelectric focusing	Isoelectric point (pI)	Protein migrates in a pH gradient until net charge is zero and it stops
2D gel electrophoresis	pI (first dimension) then mass (second)	Isoelectric focusing then SDS-PAGE; resolves thousands of spots (O'Farrell, 1975)
Western blot	Presence and size of one specific protein	SDS-PAGE, transfer to a membrane, probe with an antibody
Size-exclusion chromatography (gel filtration)	Size (hydrodynamic radius)	Large molecules elute first because they cannot enter the bead pores; checks oligomer state
Ion-exchange chromatography	Net charge at a chosen pH	Cation exchanger (negative beads) binds proteins below their pI; elute with salt or pH change
Affinity chromatography	Specific binding to an immobilized ligand	Antibody, substrate or tag on the beads; often the single most powerful purification step
His-tag purification (IMAC)	Binding of a run of histidines to a metal	Ni²⁺ (or Co²⁺) on the resin binds the His₆ tag; elute with imidazole
Hydrophobic interaction and reverse-phase chromatography	Surface or overall hydrophobicity	Reverse-phase HPLC (organic solvent) denatures proteins; good for peptides and PTH-amino acids
Salting out (ammonium sulfate precipitation)	Solubility	High salt competes for water, so proteins precipitate in order of solubility; usually the first crude step
Dialysis	Removes small molecules and changes buffer	Semi-permeable membrane holds the protein; small solutes diffuse out; used to remove urea in Anfinsen-style refolding
Ultracentrifugation	Sedimentation coefficient (svedberg units, S) and mass	Analytical runs give size and shape; S values are not additive (ribosome 70S = 50S + 30S)
Density-gradient centrifugation	Buoyant density	Sucrose or CsCl gradients; used to purify virus particles
Recombinant expression	Producing protein from a cloned gene	E. coli, yeast, insect or mammalian cells; HA needs eukaryotic cells (insect or mammalian) for glycosylation
Site-directed mutagenesis	Effect of changing one chosen residue	Swap one codon, then measure binding or stability; how roles of RBS residues are proven
UV absorbance at 280 nm	Protein concentration	Mostly Trp and Tyr (plus disulfides); Beer–Lambert A = εcl; nucleic acids absorb at 260 nm, so A260/A280 checks purity
Colorimetric assays (Bradford, BCA, biuret, Lowry)	Protein concentration	Bradford: Coomassie dye binds Arg and aromatic residues and shifts to blue; biuret detects peptide bonds with Cu²⁺
Circular dichroism (far-UV)	Secondary structure content	α-helix: minima at 208 and 222 nm; β-sheet: minimum near 218 nm; follows unfolding with temperature or denaturant
Intrinsic fluorescence	Environment of Trp (and Tyr)	Buried Trp emits at shorter wavelength (blue-shifted); exposure on unfolding shifts emission to about 350 nm
FRET	Distance between two fluorophores (about 1–10 nm)	Energy transfer falls with the sixth power of distance; a "molecular ruler" for conformational change
Atomic absorption spectroscopy	Amount of a metal ion	Measures metal content, such as Zn²⁺ in a zinc-binding protein
Chemical denaturation curves	Stability (ΔG of unfolding)	Titrate urea or guanidinium chloride, follow CD or fluorescence; midpoint and slope give ΔG
Differential scanning calorimetry (DSC)	Heat absorbed on unfolding; melting temperature Tm	Peak at Tm; area gives the enthalpy of unfolding
Thermal shift assay (differential scanning fluorimetry)	Tm by dye fluorescence	Dye binds exposed hydrophobic core as protein melts; ligand binding raises Tm; high throughput
Isothermal titration calorimetry (ITC)	Heat of binding	Gives Kd, stoichiometry and ΔH in one experiment, and hence ΔG and TΔS
Surface plasmon resonance (SPR)	Binding in real time without labels	One partner on a gold chip; gives on-rate, off-rate and Kd
Equilibrium dialysis	Binding affinity of a small ligand	Ligand passes the membrane, protein cannot; excess ligand on the protein side measures binding
Enzyme activity assays	Rate of product formation	Often followed by a color or absorbance change (NADH at 340 nm); define units of activity
Hemagglutination (HA) assay	Amount of virus (HA titer)	Virus cross-links red cells into a lattice instead of a button; titer = highest dilution that still agglutinates (Hirst, 1941)
Hemagglutination inhibition (HAI) assay	Antibodies that block receptor binding	Serum + virus, then red cells: antibodies prevent agglutination; titer ≥ 40 is the usual correlate of protection
Microneutralization assay	Antibodies that stop infection of cells	Measures neutralization by any mechanism, including stem antibodies that HAI misses
Plaque assay and TCID₅₀	Infectious virus titer	Counts plaques or infected wells; HA assay counts particles, not infectivity
ELISA	Amount of antigen or antibody	Antibody–enzyme conjugate makes a colored product; sandwich or indirect formats
Glycan arrays	Which glycans a protein binds	Hundreds of glycans printed on a slide; how α2,3 versus α2,6 preference is measured (used for 3UBE's virus)
Antigenic cartography	Antigenic distance between flu strains	Maps HAI data onto a 2D map; guides vaccine strain choice
Deep mutational scanning	Effect of every single mutation	Libraries of all variants selected for function or escape; maps HA's tolerance to mutation
DSSP	Secondary structure assigned from 3D coordinates	Uses backbone H-bond patterns and geometry; assigns, does not predict (Kabsch and Sander, 1983)
Sequence alignment and BLAST	Similarity between sequences	BLAST searches a database for similar sequences; scored with BLOSUM or PAM matrices; reports E-values
Multiple sequence alignment	Conserved and variable positions across many sequences	Conserved residues are structurally or functionally important; covarying pairs hint at contacts
Hydropathy plot (Kyte–Doolittle)	Hydrophobicity along a sequence	Windows of about 19–20 hydrophobic residues predict transmembrane helices
Chou–Fasman and later secondary-structure prediction	Helix, strand or coil per residue from sequence	Uses residue propensities (Ala, Leu, Glu, Met favor helix; Val, Ile, Tyr favor strand; Pro, Gly break)
Homology (comparative) modeling	3D model built on a related known structure	Needs a template, roughly above 30% sequence identity; SWISS-MODEL
Threading (fold recognition)	Which known fold a sequence fits	Works when sequence identity is too low for homology modeling
Molecular dynamics	Atomic motion over time	Force field of bonds, angles, torsions, van der Waals and electrostatics; nanoseconds to microseconds
Docking	Likely pose and strength of ligand binding	Scores complementary shape and interactions; used for drug and glycan placement
Rosetta and de novo design	New sequences for a chosen shape or binder	Baker lab; designed HA-stem binders and nanoparticle vaccines
AlphaFold and pLDDT	Predicted structure from sequence; per-residue confidence	pLDDT above 90 very high, below 50 often disordered; PAE gives domain placement confidence; a prediction, not a measurement
Ramachandran validation	Whether φ/ψ angles are in allowed regions	Good structures have about 98% of residues in favored regions; outliers flag errors (MolProbity)
RMSD	Average distance between matched atoms after superposition	Under 1 Å nearly identical; 1–2 Å same fold; squares make large deviations dominate
Contact map and distance matrix	Which residue pairs are close	Symmetric; helices as bands by the diagonal, sheets as off-diagonal lines; rotation-invariant

The single most common trap in this table is mixing up what each method separates by: SDS-PAGE separates by size alone, native PAGE by size, shape and charge, isoelectric focusing by pI, ion exchange by charge, size-exclusion by size (largest out first). The second is confusing assignment with prediction: DSSP assigns secondary structure from coordinates, while Chou–Fasman and AlphaFold predict from sequence.

People and milestones

History questions reward knowing a short list of names with one fact each. Most fall into three groups: those who solved the first structures and sequences, those who explained folding, and those who built the methods (often with a Nobel Prize). A few are specific to influenza and to 3UBE itself. The year given is the year of the discovery; the Nobel year, if any, is listed with the field.

ReferencePeople and milestones in protein science· 67 rows

Nobel fields: Chemistry or Physiology or Medicine ("Medicine" below), with the year awarded.

67 rows

Person or team	Year	Contribution	Why it is asked
Gerardus Mulder (name suggested by Jöns Jacob Berzelius)	1838	Coined "protein", from Greek "of first importance"	Origin of the word
Emil Fischer	1894 (enzyme model); 1902 Nobel Chemistry (for sugars and purines)	Lock-and-key model of enzyme specificity; also showed proteins are chains of amino acids joined by peptide bonds	Lock and key versus induced fit
Max von Laue; W. H. and W. L. Bragg	1912–1913; Nobel Physics 1914 (Laue), 1915 (Braggs)	X-ray diffraction by crystals; Bragg's law nλ = 2d sin θ	Foundation of crystallography
Frederick Banting and Charles Best (with J. J. R. Macleod)	1921; Nobel Medicine 1923 (Banting and Macleod)	Isolated insulin and treated diabetes	First protein hormone used as a drug
James B. Sumner	1926; Nobel Chemistry 1946 (shared with Northrop and Stanley)	Crystallized urease	Proved enzymes are proteins
The Svedberg	1920s; Nobel Chemistry 1926	Analytical ultracentrifuge	Showed proteins have defined masses; the svedberg unit (S)
Wilson Smith, Christopher Andrewes and Patrick Laidlaw	1933	Isolated human influenza virus by infecting ferrets	Influenza A shown to be a virus; ferrets remain the standard model
Max Bergmann and Leonidas Zervas	1932–1936	Carbobenzoxy protecting group; an early stepwise peptide degradation	Precursor to Edman's method, which replaced it
Arne Tiselius	1937; Nobel Chemistry 1948	Electrophoresis of serum proteins	Origin of separating proteins by charge in an electric field
George Hirst	1941	Discovered that influenza virus agglutinates red cells; built the HA and HAI assays	Why the protein is called hemagglutinin; HAI still measures flu immunity
Archer Martin and Richard Synge	1941; Nobel Chemistry 1952	Partition chromatography	Basis of amino acid analysis and HPLC
Dorothy Crowfoot Hodgkin	Penicillin 1945, vitamin B12 mid-1950s, insulin 1969; Nobel Chemistry 1964	X-ray structures of important biomolecules	The 1964 Nobel was for penicillin and B12; insulin came five years after the prize
Linus Pauling (with Itano, Singer and Wells)	1949	Showed sickle-cell hemoglobin has an abnormal charge	First "molecular disease"
Frederick Sanger	Insulin sequence complete 1955; Nobel Chemistry 1958 and 1980	First complete protein sequence (insulin; FDNB labeling of N-termini, overlapping fragments); later DNA sequencing	Proved a protein has one exact sequence; only person with two Chemistry Nobels
Pehr Edman	1950; automated sequenator with Geoffrey Begg 1967	Stepwise N-terminal degradation with phenyl isothiocyanate	Sequencing without destroying the rest of the chain
Linus Pauling and Robert Corey (with Herman Branson)	1951; Pauling's Nobel Chemistry 1954	Predicted the α helix and β pleated sheet from model building with planar peptide bonds	Secondary structure predicted before any protein structure was solved
Rosalind Franklin and Raymond Gosling	1952	Photo 51, the X-ray fiber diffraction pattern of B-DNA	Context for X-ray history; Watson, Crick and Wilkins received the 1962 Nobel Medicine (Franklin had died in 1958)
Frank Macfarlane Burnet	1957; Nobel Medicine 1960 (with Medawar, for acquired tolerance)	Clonal selection theory	How each B cell makes one antibody and is selected by antigen
Alick Isaacs and Jean Lindenmann	1957	Discovered interferon, using influenza virus	Innate antiviral response; NS1 blocks it
John Kendrew	1958 (6 Å), 1960 (2 Å); Nobel Chemistry 1962	Myoglobin, the first protein structure	Mostly α helix; confirmed Pauling's helix
Max Perutz	1959–1960; Nobel Chemistry 1962	Hemoglobin structure; used heavy-atom isomorphous replacement to solve phases	Quaternary structure; how O₂ binding shifts subunits
Daniel Koshland	1958	Induced-fit model	Enzyme and binding site shape changes on binding
Walter Kauzmann	1959	Identified the hydrophobic effect as a main driving force of folding	Folding is driven by water entropy
Christian Anfinsen	Ribonuclease refolding late 1950s–1961; Nobel Chemistry 1972 (with Moore and Stein)	Denatured RNase A (urea + β-mercaptoethanol) refolds and regains activity when they are removed slowly	Anfinsen's dogma: sequence determines the native structure
Stanford Moore and William Stein	1950s–1960s; Nobel Chemistry 1972	Automated amino acid analysis; RNase A sequence and active-site chemistry	Shared Anfinsen's prize
G. N. Ramachandran (with Ramakrishnan and Sasisekharan)	1963	Plot of allowed backbone φ/ψ angles from steric clashes	Ramachandran plot; still used to validate structures
David Phillips	1965	Lysozyme, the first enzyme structure (2 Å)	First structure that explained catalysis
Margaret Dayhoff	1965	Atlas of Protein Sequence and Structure; one-letter amino acid code; PAM matrices	Origins of bioinformatics and sequence databases
Jacques Monod, Jeffries Wyman and Jean-Pierre Changeux	1965	Concerted (MWC) model of allostery	T and R states for hemoglobin and allosteric enzymes
Cyrus Levinthal	1968–1969	Levinthal's paradox: a random search of conformations would take longer than the age of the universe	Folding must follow pathways or a funnel, not random search
Bruce Merrifield	1963; Nobel Chemistry 1984	Solid-phase peptide synthesis	Chemical synthesis of peptides on beads
Rodney Porter and Gerald Edelman	1959–1969; Nobel Medicine 1972	Structure of antibodies (heavy and light chains, Fab and Fc)	Antibody architecture
Ulrich Laemmli	1970	Discontinuous SDS-PAGE system	The standard protein gel
Protein Data Bank founded	1971	Archive at Brookhaven with 7 structures	Now run by wwPDB; over 250,000 entries
Günter Blobel	Signal hypothesis 1971, shown 1975; Nobel Medicine 1999	Proteins carry intrinsic signals that direct them to their location (signal peptides)	Why HA's signal peptide sends it into the ER
Patrick O'Farrell	1975	High-resolution 2D gel electrophoresis	IEF plus SDS-PAGE
Richard Henderson and Nigel Unwin	1975 (7 Å); atomic model 1990	Bacteriorhodopsin by electron crystallography; first view of transmembrane helices	Start of cryo-EM; Henderson shared the 2017 prize
Georges Köhler and César Milstein (with Niels Jerne)	1975; Nobel Medicine 1984	Hybridomas for monoclonal antibodies	Monoclonal antibody tools and drugs
Erwin Neher and Bert Sakmann	Patch clamp 1976; Nobel Medicine 1991	Recorded currents through single ion channels	Channels open and close as single molecules
Susumu Tonegawa	1976; Nobel Medicine 1987	Gene rearrangement creates antibody diversity	V(D)J recombination
Ian Wilson, John Skehel and Don Wiley	1981	First HA structure (H3, 3 Å), from bromelain-cleaved HA trimer	Defined HA1 head, HA2 stem, receptor site and antigenic sites
Jane Richardson	1981	Ribbon diagrams and a taxonomy of protein folds	The cartoon representation used in JUDE
Stanley Prusiner	Named prions 1982; Nobel Medicine 1997	Infectious proteins (PrP) cause scrapie and CJD	Protein-only infection; templated misfolding
Gary Rogers and James Paulson	1983	Human flu isolates prefer α2,6-linked sialic acid, avian ones α2,3	Basis of receptor specificity and host range
Jack Kyte and Russell Doolittle	1982	Hydropathy scale and plot	Predicting transmembrane segments
Wolfgang Kabsch and Chris Sander	1983	DSSP algorithm	How viewers and the PDB assign helices and sheets
Kurt Wüthrich	First protein structure by NMR 1985; Nobel Chemistry 2002	Method to determine protein structures in solution by NMR	Shared 2002 with John Fenn and Koichi Tanaka (ESI and MALDI mass spectrometry)
Joseph Bryngelson and Peter Wolynes	1987	Energy landscape theory and the principle of minimal frustration	Folding funnel
Edwin Krebs and Edmond Fischer	1950s work; Nobel Medicine 1992	Reversible protein phosphorylation (glycogen phosphorylase)	Kinases and phosphatases as switches
Stephen Altschul and colleagues	1990	BLAST	Fast database similarity searching
Peter Agre and Roderick MacKinnon	Aquaporin 1992; K⁺ channel structure 1998; Nobel Chemistry 2003	Water channels; structure and selectivity of ion channels	Membrane channel structure and selectivity
Paul Boyer, John Walker and Jens Skou	Nobel Chemistry 1997	ATP synthase mechanism and structure (Boyer, Walker); Na⁺/K⁺-ATPase (Skou)	Rotary catalysis; ion pumps
Per Bullough, Fred Hughson, John Skehel and Don Wiley	1994	Structure of HA at fusion pH	Huge refolding: fusion peptide moves about 100 Å; spring-loaded mechanism
John Moult and colleagues	1994	Started CASP, the blind prediction competition	AlphaFold2's 2020 result was judged at CASP14
Osamu Shimomura, Martin Chalfie and Roger Tsien	GFP discovered 1962; Nobel Chemistry 2008	Discovery and development of GFP	Fluorescent protein tags
Aaron Ciechanover, Avram Hershko and Irwin Rose	Nobel Chemistry 2004	Ubiquitin-mediated protein degradation	Proteins are tagged for destruction
Brian Kuhlman, David Baker and colleagues	2003	Top7, a designed protein with a fold not seen in nature	Beginning of de novo protein design
Jeffery Taubenberger, Terrence Tumpey and colleagues	2005	Sequenced and reconstructed the 1918 pandemic virus	Pandemic history; HA receptor changes
Roger Kornberg	Nobel Chemistry 2006	Structural basis of eukaryotic transcription (RNA polymerase II)	Protein machines that read DNA
Venkatraman Ramakrishnan, Thomas Steitz and Ada Yonath	Nobel Chemistry 2009	Ribosome structure and function	The ribosome is a ribozyme
Robert Lefkowitz and Brian Kobilka	Nobel Chemistry 2012	G-protein-coupled receptors	Rhodopsin family; 7 transmembrane helices
Rui Xu, Ryan McBride, Corwin Nycholat, James Paulson and Ian Wilson	2012 (J. Virol. 86:982–990; entry released 2011)	Crystal structures of 2009 pandemic HA with receptor analogs, including 3UBE with LSTc	The structure named in the 2027 rules
Jacques Dubochet, Joachim Frank and Richard Henderson	Nobel Chemistry 2017	Cryo-electron microscopy for high-resolution structures in solution	Vitrification (Dubochet), image averaging (Frank), atomic resolution (Henderson)
Frances Arnold; George Smith and Gregory Winter	Nobel Chemistry 2018	Directed evolution of enzymes; phage display of peptides and antibodies	Evolution as a protein engineering tool
John Jumper, Demis Hassabis and the DeepMind team	AlphaFold2 at CASP14 2020; Nobel Chemistry 2024	Protein structure prediction from sequence; database of over 200 million predictions	Half of the 2024 prize
David Baker	Nobel Chemistry 2024	Computational protein design (Rosetta)	Other half of the 2024 prize; designed binders and vaccines
Katalin Karikó and Drew Weissman	Nobel Medicine 2023	Nucleoside-modified mRNA, enabling mRNA vaccines	Vaccine platform now being tested for influenza

Two attribution traps recur. The 2024 Chemistry prize was split: half to Baker for designing proteins, half to Hassabis and Jumper for predicting them. And Hodgkin's 1964 prize predates her insulin structure; Sanger, not Hodgkin, sequenced insulin. When a question asks who solved the "first" structure, the answer is Kendrew (myoglobin); the first enzyme structure is Phillips (lysozyme); the first NMR structure is Wüthrich's group; the first HA structure is Wilson, Skehel and Wiley.

Condensed facts

Everything below fits on one 8.5 × 11 in sheet, printed both sides. It deliberately repeats the guide's most quotable facts in compressed form: the chapters explain them, the Breadth reference tables hold the long tail, and this is what goes into the event with you. Print it, then add your own 3UBE helix and strand boundaries and your legend's color key.

Amino acids

Charge math. Fraction protonated = 1/(1 + 10^(pH − pKₐ)). Acidic group: charge −1/(1 + 10^(pKₐ − pH)). Basic group: +1/(1 + 10^(pH − pKₐ)). pI = average of the two pKₐ values flanking the neutral form (acidic side chain → two lowest; basic → two highest). Below pI net +, above pI net −. Peptide: one N-terminus (+, pKₐ ≈ 9) and one C-terminus (−, pKₐ ≈ 2–3) plus ionizable side chains.

Mass. Residue ≈ 110 Da average; n ≈ M/110. Chain mass = Σ residue masses + 18 Da. Condensation releases one water per bond.

Structure and folding

Levels. 1° sequence (peptide bonds) → 2° local backbone shape (backbone H-bonds) → 3° whole fold of one chain (side-chain forces, disulfides) → 4° several chains (HA = trimer). Denaturation destroys 2°/3°/4°, never 1°.

Thermodynamics. ΔG = ΔH − TΔS; folded state favored by only ≈ 20–60 kJ/mol. Tm = half unfolded, ΔG = 0. Anfinsen: sequence determines fold. Levinthal: folding cannot be a random search → funnel landscape. Chaperones (Hsp60, Hsp70) prevent wrong hydrophobic contacts. Denaturants: heat, extremes of pH (breaks ionic bonds), urea and guanidinium (H-bonds and hydrophobic core), detergents like SDS, organic solvents, heavy metals (bind thiols), reducing agents (disulfides only).

Motifs. β-α-β, Greek key, β-barrel, Rossmann fold, coiled coil (leucine zipper), helix-turn-helix, zinc finger (Cys₂His₂ + Zn²⁺), EF hand. Globular (soluble, hydrophobic core) vs fibrous (collagen, keratin) vs membrane (hydrophobic outside, ≈ 20-residue transmembrane helices).

Function, mutation and methods

• Enzymes lower activation energy by stabilizing the transition state; they do not change ΔG or Keq. Induced fit beats lock-and-key. Cofactors/coenzymes assist; some enzymes are RNA (ribozymes).
• Inhibition: competitive (binds the active site, overcome by more substrate), noncompetitive/allosteric (binds elsewhere, not overcome), uncompetitive (binds ES), irreversible (covalent).
• Mutations: silent, missense (conservative vs nonconservative), nonsense, frameshift (insertion/deletion not in threes). Surface change ≈ tolerated; core change destabilizes; active-site change alters binding. Sickle cell = Glu6 → Val in β-globin (polar → hydrophobic patch).
• Central dogma: DNA → RNA → protein; reverse transcription is the exception. Codon = 3 bases; 64 codons, 20 amino acids; AUG start; UAA/UAG/UGA stop.
• Structure methods: X-ray (needs crystals, atomic resolution, most PDB entries, gives electron density; R-factor and resolution in Å), NMR (solution, small proteins, gives an ensemble), cryo-EM (large complexes, no crystal needed), AlphaFold (prediction from sequence, pLDDT confidence — not a measurement). PDB file: ATOM, HETATM, HELIX, SHEET, SSBOND records; coordinates in Å.

JUDE / Jmol

Influenza and immunity

• Virion: enveloped, negative-sense ssRNA, 8 segments, Orthomyxoviridae; HA (binds sialic acid, fusion), NA (cuts sialic acid), M2 (proton channel), M1 matrix, NP, polymerase PB1/PB2/PA, NS1 (blocks interferon).
• Cycle: attachment to sialic acid → endocytosis → endosome acidifies (pH ≈ 5–6) → M2 lets protons in → HA refolds, fusion peptide inserted, membranes merge → vRNPs to the nucleus → cap-snatching transcription → new proteins and genomes → budding → NA frees progeny.
• HA0 must be cleaved to HA1 + HA2 (disulfide-linked) by a host protease before the virus is infectious; one basic residue (airway proteases) vs a polybasic site (furin, highly pathogenic avian strains).
• Drift = point mutations in HA/NA antigenic sites → annual vaccine updates. Shift = reassortment of whole segments → new HA most people lack immunity to.
• Receptors: α2,3 = avian-type (human lower airway), α2,6 = human-type (upper airway). Pigs carry both. H1 specificity switch: Asp190 + Asp225; H2/H3: Q226L + G228S. D225G (D222G in H1 numbering) shifts toward α2,3. Single contact is weak (mM); binding works by avidity; HA–NA balance must match.
• Binding site: floor Tyr98, Trp153, His183, Tyr195; rim 130-loop, 150-loop, 190-helix, 220-loop. Carboxylate held by the 130-loop and Gln226; glycerol by Tyr98/His183; N-acetyl methyl on Trp153.
• Innate: mucus and cilia, RIG-I and TLRs sensing viral RNA, type I interferon, ISGs, NK cells, macrophages, dendritic cells, inflammation (fever and aches come from cytokines).
• Adaptive: dendritic cells present antigen; MHC I → CD8⁺ cytotoxic T cells kill infected cells; MHC II → CD4⁺ helper T cells help B cells; B cells → plasma cells → antibodies (IgG in blood, IgA on mucosa, IgM first); class switching and affinity maturation; memory B and T cells.
• Antibodies to the HA head neutralize but drift; the stem is conserved (universal vaccine target). HAI assay measures antibodies that block HA from agglutinating red cells.
• Drugs: oseltamivir/zanamivir/peramivir → NA; baloxavir → polymerase (PA endonuclease, cap-snatching); amantadine/rimantadine → M2 (widespread resistance). Vaccines: inactivated, recombinant HA, live attenuated (nasal, mucosal IgA); strains chosen each year.

References and further reading

• Xu R et al., "Structural characterization of the hemagglutinin receptor specificity from the 2009 H1N1 influenza pandemic", Journal of Virology 2012, the article named in the 2027 rules: https://doi.org/10.1128/JVI.06322-11
• RCSB Protein Data Bank entry 3UBE (coordinates, ligands and structure summary): https://www.rcsb.org/structure/3UBE
• 3D Molecular Designs, Protein Modeling event page (the modified structure file and JUDE): https://learn.3dmoleculardesigns.com/science-olympiad-protein-modeling-event
• JUDE (Jmol User Design Environment), free in a browser: https://learn.3dmoleculardesigns.com/digital-modeling-hub/jude_online_jmol
• Xu R et al. 2012, free full text at PubMed Central (figures of the LSTc and LSTa complexes): https://pmc.ncbi.nlm.nih.gov/articles/PMC3255799/
• RCSB PDB "Molecule of the Month": Hemagglutinin, a plain-language tour of the protein you are modeling: https://pdb101.rcsb.org/motm/76
• Jmol documentation, the scripting commands JUDE runs underneath: https://chemapps.stolaf.edu/jmol/docs/
• Berg, Tymoczko and Stryer, Biochemistry, for side-chain chemistry, folding and enzyme mechanisms, latest edition
• CDC, influenza antiviral drug summary for clinicians: https://www.cdc.gov/flu/hcp/antivirals/summary-clinicians.html
• WHO, influenza vaccine composition recommendations (how each season's strains are chosen): https://www.who.int/teams/global-influenza-programme/vaccines/who-recommendations
• NCBI Bookshelf, Janeway's Immunobiology, for antigen presentation, antibodies and immune memory: https://www.ncbi.nlm.nih.gov/books/NBK27090/
• Skehel JJ and Wiley DC, "Receptor binding and membrane fusion in virus entry: the influenza hemagglutinin", Annual Review of Biochemistry 2000: https://doi.org/10.1146/annurev.biochem.69.1.531
• CDC, "How influenza viruses change: drift and shift": https://www.cdc.gov/flu/php/viruses/change.html
• NIAID, influenza overview and immune response: https://www.niaid.nih.gov/diseases-conditions/influenza
• Anfinsen CB, "Principles that govern the folding of protein chains", Science 1973: https://doi.org/10.1126/science.181.4096.223
• OpenStax, Biology 2e, chapters on biological macromolecules and on viruses and immunity: https://openstax.org/details/books/biology-2e
• Lehninger, Principles of Biochemistry, for amino-acid properties, pKₐ values and folding thermodynamics, latest edition
• Science Olympiad, Protein Modeling event page and rules: https://www.soinc.org
• SciOly Wiki, Protein Modeling: https://scioly.org/wiki/index.php/Protein_Modeling
• Practice more: the Scio.ly test archive has public Protein Modeling tests you can filter by subtopic.