Glossary
Every term the course uses, in a sentence or two, under the lesson that teaches it. In the lessons, a dotted underline marks a term defined here; point at it to see the definition.
0.1 Cells, proteins, and scale
- micrometer
- 10⁻⁶ m. Cells are 1–100 µm across.
- organelle
- A membrane-enclosed compartment inside a eukaryotic cell, such as the nucleus or a mitochondrion.
- eukaryote
- An organism whose cells keep their DNA in a nucleus and have other membrane-enclosed compartments: animals, plants, and fungi.
- prokaryote
- A bacterium or archaeon, whose cells lack a membrane-enclosed nucleus.
- cytoplasm
- Everything inside the cell membrane (in a eukaryote, outside the nucleus).
- cell membrane
- The lipid bilayer and its proteins that enclose a cell and control what enters and leaves.
- cytosol
- The watery part of the cytoplasm, crowded with proteins, salts, and small molecules. It is reducing, so disulfide bonds are rare there.
- phospholipid
- A lipid with a charged, water-loving head and two oily fatty-acid tails; the main component of membranes.
- lipid bilayer
- Two layers of lipids with their nonpolar tails facing inward and polar heads facing water; a common membrane arrangement about 4 nm thick.
- enzyme
- A protein (occasionally an RNA) that catalyzes a particular reaction.
- catalyst
- Something that speeds up a reaction and comes out unchanged.
- receptor
- A protein that binds a signal molecule, such as a hormone, and changes shape in response.
- nanometer
- 10⁻⁹ m. A typical protein is a few nanometers across.
- ångström
- 10⁻¹⁰ m, or 0.1 nm. Bond lengths are 1–2 Å, which is why structure files measure distance in ångströms.
- dalton
- The unit of molecular mass, one-twelfth of a carbon-12 atom (about one hydrogen atom). A molecule of M daltons weighs M grams per mole.
- kilodalton
- 1,000 daltons. A residue averages about 110 Da, so 30 kDa is about 270 residues.
- antibody
- An immune-system protein that binds a foreign molecule tightly and marks it for destruction.
- cell nucleus
- The organelle that holds most of a eukaryotic cell's DNA.
- extracellular
- Outside the cell, where secreted proteins often carry disulfide bonds.
- hormone
- A signal molecule, some of them proteins like insulin, carried through the body to cells with the matching receptor.
- membrane protein
- A protein embedded in or associated with a membrane. Regions facing lipid tails are usually nonpolar; a water-filled pore can have a polar lining.
- transporter
- A membrane protein that lets a particular molecule or ion cross the membrane.
0.2 Life as information processing
- gene
- A stretch of DNA that is transcribed into RNA, usually one that encodes a protein.
- polymer
- A molecule built by linking many smaller units, called monomers.
- monomer
- A small molecular building block that can be joined into a polymer, such as a nucleotide or amino acid.
- DNA
- Deoxyribonucleic acid: a double-stranded nucleotide polymer that stores genetic information as a string over A, C, G, and T.
- RNA
- Ribonucleic acid: a usually single-stranded nucleotide polymer over A, C, G, and U. mRNA, tRNA, and rRNA carry out protein synthesis.
- nucleotide
- The monomer of DNA and RNA, made of a base, a five-carbon sugar, and a phosphate group.
- purine
- A base with two fused rings, adenine (A) or guanine (G).
- pyrimidine
- A base with a single ring, cytosine (C), thymine (T), or uracil (U).
- complementary
- Paired by Watson–Crick rules: A with T (or U), G with C.
- base pair
- Two bases on opposite strands joined by hydrogen bonds; also the unit of length for double-stranded DNA.
- transcription
- Copying DNA into RNA, done by RNA polymerase.
- template strand
- The DNA strand RNA polymerase reads; the RNA is complementary to it.
- coding strand
- The DNA strand whose sequence matches the RNA, with T for U.
- intron
- Part of a gene cut out of the RNA before translation.
- exon
- Part of a gene kept in the final mRNA.
- mRNA
- Messenger RNA: the copy of a protein-coding gene that the ribosome reads.
- ribosome
- A complex of RNA and proteins that reads mRNA and joins amino acids into a protein chain.
- tRNA
- Transfer RNA: a small folded RNA that carries one kind of amino acid and pairs with the matching mRNA codon.
- translation
- Building a protein from an mRNA's codons, done by the ribosome with tRNAs as adaptors.
- codon
- Three consecutive bases on an mRNA that specify one amino acid or a stop.
- genetic code
- The table mapping the 64 codons to the 20 amino acids and stop; nearly identical in all organisms.
- reading frame
- One of three ways to split a sequence into codons, set by the starting base.
- start codon
- A codon used to initiate translation and establish a reading frame, usually AUG. Cellular signals and surrounding sequence help determine which start is used.
- synonymous mutation
- A codon change that leaves the encoded amino acid unchanged. It can still affect RNA processing or translation.
- central dogma
- Information flows from DNA to RNA (transcription) and from RNA to protein (translation).
- nucleic acid
- DNA or RNA, a polymer of nucleotides.
- base (of a nucleotide)
- The variable ring-shaped part of a nucleotide (A, C, G, T, or U), which carries the information. Bases on opposite strands pair by hydrogen bonds.
- stop codon
- UAA, UAG, or UGA: ends translation and releases the chain.
0.3 Three ways to write down a molecule
- PDB
- The Protein Data Bank, an archive of experimentally determined molecular structures and associated information. An identifier such as 1UBQ selects a deposited entry.
- invariant
- Unchanged by a specified transformation; for example, pairwise distances under a rigid rotation.
- rigid motion
- Rotating and translating a whole molecule: every coordinate changes, no distance does, so the molecule is unchanged.
- equivariant
- Changing consistently with a transformation of the input; for example, predicted coordinates rotating with the input structure.
- structural bioinformatics
- The part of bioinformatics that represents, stores, analyzes, and predicts the 3D structures of biomolecules.
- structure determination
- Building an atomic model from experimental data such as X-ray diffraction, NMR, or cryo-EM.
- structure prediction
- Computing a molecule's 3D coordinates from its sequence, without an experiment.
- crystal structure
- A structure determined by X-ray diffraction from a crystal.
0.4 Reading molecule drawings
- methyl
- –CH₃, the smallest carbon group; nonpolar. Alanine's whole side chain.
- molecular formula
- A count of a molecule's atoms, such as C₂H₆O, saying nothing about how they connect.
- structural formula
- A drawing of a molecule showing every atom and every bond.
- condensed formula
- Each carbon written with its hydrogens, as in CH₃CH₂OH.
- skeletal formula
- A bond drawing with carbon symbols and their hydrogens mostly omitted. Unlabeled corners and endpoints are carbons; a labeled endpoint is the atom named there.
- lone pair
- A pair of valence electrons not assigned to a bond. Ordinary neutral alcohol oxygen has two lone pairs; an amine nitrogen usually has one.
- amino group
- –NH₂. At pH 7 it usually gains an H to become ammonium, –NH₃⁺.
- formal charge
- Charge assigned by giving an atom all its unshared electrons and half its bonding electrons, then comparing with the neutral atom's valence count. It differs from a force-field partial charge.
- valence
- An atom's combining capacity in a specified chemical state. Common neutral groups have H making one bond, O two, N three, and C four, with double bonds counting twice.
- carboxyl
- –COOH. At pH 7 it loses the H to become carboxylate, –COO⁻.
- functional group
- A small group of atoms with characteristic chemistry, such as hydroxyl or carboxyl, which it lends to any molecule carrying it.
- hydroxyl
- –OH: polar, donates and accepts hydrogen bonds. Found on serine, threonine, and tyrosine.
- carbonyl
- C=O. Every peptide bond has one.
- amide
- –C(=O)–NH–, a carbonyl bonded to a nitrogen. The peptide bond is an amide.
- thiol
- –SH, the tip of cysteine's side chain. Two can join as a disulfide, –S–S–.
- phenyl
- A benzene ring attached by one carbon, –C₆H₅. Phenylalanine's side chain is CH₂ plus phenyl.
- aromatic
- A flat ring, like benzene, whose double-bond electrons are shared around it; drawn with alternating double bonds or a circle.
- phosphate
- A group containing phosphorus bonded to four oxygens. Its attachments and protonation set its charge; phosphate groups link nucleic-acid backbones and can modify protein behavior.
- R group
- "The rest of the molecule" in a drawing; in an amino acid, the side chain.
- wedge bond
- A solid wedge is a bond toward the viewer, a dashed wedge one pointing away; plain lines lie in the page.
Lab 0 A first look at a PDB file
- HETATM
- The PDB record type for atoms outside standard residues, such as water, ligands, metal ions, and modified residues.
- radius of gyration
- The root-mean-square distance of atoms from their center: one number for size and compactness.
- heavy atom
- Any atom other than hydrogen. X-ray structures usually list only heavy atoms.
1.1 Atoms, electrons, and covalent bonds
- ion
- An atom or group with a whole-number net charge from gained or lost electrons (or protons).
- kT
- The thermal energy scale kBT for one molecule, or RT per mole: about 0.6 kcal/mol at room temperature. Comparing an energy or barrier with this scale helps assess populations or rates.
- atomic number
- The number of protons in the nucleus, which sets the element.
- isotope
- A form of an element with a particular number of neutrons; chemically almost identical to the others.
- cation
- A positively charged ion.
- anion
- A negatively charged ion.
- orbital
- The region around a nucleus where a given electron is likely to be found. Each holds at most two electrons.
- covalent bond
- A chemical bond involving shared electron density. Ordinary covalent bonds remain intact during protein folding; changing them requires a chemical reaction.
- double bond
- A bond represented by two shared electron pairs. A C=C bond is shorter than a C–C single bond and has a much larger rotational barrier.
- octet rule
- Atoms such as C, N, and O bond until they have eight outer electrons; hydrogen needs two.
- kcal/mol
- Kilocalories per mole, the usual unit for molecular energies. 1 kcal = 4.184 kJ; a mole is 6.02 × 10²³ molecules.
- bond length
- The distance between two bonded atoms' centers; nearly constant for a given pair and bond type (C–C, 1.53 Å).
- conformation
- One 3D arrangement of a molecule, reachable from others by rotating about single bonds.
- isomer
- Molecules with the same atoms arranged differently: structural isomers differ in bond graph, cis–trans and optical isomers in 3D arrangement.
- resonance
- Electrons that no single drawing of single and double bonds captures, so the real bonds are in between, as in a carboxylate or the peptide group.
- covalent radius
- An atom's share of a typical bond length. Atoms closer than the sum of their covalent radii, plus a tolerance, are taken as bonded.
- valence shell
- The outermost electron shell, whose electrons form bonds.
1.2 Electronegativity and polarity
- electronegativity
- How strongly an atom pulls shared electrons. O and N pull much harder than C and H.
- partial charge
- A fraction of an electron's charge on an atom in a polar bond (δ⁺ or δ⁻). Force fields assign one to every atom.
- polar
- Sharing electrons unevenly, leaving partial positive and negative ends. Polar groups attract water and each other.
- force field
- A set of formulas and fitted parameters that computes a molecule's energy from its atom positions, used in simulation, minimization, and docking.
- dipole moment
- The vector sum of charges times positions: how far apart a molecule's positive and negative ends are.
- nonpolar
- Having little charge separation, as in many hydrocarbon groups. Nonpolar groups still interact with water, including through dispersion forces.
- ligand
- Any small molecule or ion that binds to a protein.
1.3 Acids, bases, pH, and pKa
- isoelectric point
- The pH at which a molecule's net charge is zero.
- pKa
- The pH at which a group is protonated half the time; below it mostly protonated, above it mostly not.
- pH
- Minus the base-10 logarithm of proton activity, approximated by concentration relative to 1 M in dilute solution. Lowering pH by one unit increases proton activity tenfold.
- protonation state
- Which protons are attached to an ionizable group. Gaining one proton raises its net charge by one; the resulting charge also depends on the starting group.
- molar
- Moles per liter, written M; 1 M is 6.02 × 10²³ molecules per liter.
- physiological pH
- About 7.2–7.4, the pH inside cells and in blood. "Charged" without qualification means charged at this pH.
- acid (proton donor)
- Gives up a proton (H⁺). A carboxyl group is neutral with its proton and negative without it.
- base (proton acceptor)
- Takes up a proton (H⁺). An amino group is neutral without it and positive with it.
- Henderson–Hasselbalch equation
- pH = pKa + log10([deprotonated]/[protonated]). It makes the protonated fraction a logistic function of pH.
1.4 Noncovalent interactions
- noncovalent interaction
- An interaction that leaves ordinary covalent connectivity unchanged, such as a hydrogen bond, ion pair, or van der Waals contact. Its energy depends on geometry and environment.
- hydrogen bond
- A directional interaction between a donor-bound hydrogen and a suitable acceptor. In proteins, donor and acceptor are commonly N or O, with favorable heavy-atom separations often around 2.7–3.2 Å.
- van der Waals force
- A weak, short-range attraction between any two atoms from fluctuating electron clouds, a fraction of a kT per contact.
- salt bridge
- An interaction between oppositely charged groups, often identified by a positive N and negative O within about 4 Å. Its contribution to stability depends on solvation and the surrounding structure.
- donor
- In a hydrogen bond, the N or O atom that holds the hydrogen.
- acceptor
- In a hydrogen bond, the N or O atom the hydrogen points at, which offers a lone pair of electrons.
- dielectric constant
- A continuum measure of electrical polarization, about 80 for water near room temperature. Low values such as 2–4 are sometimes used for protein interiors, whose effective response depends on the model.
- solvation
- Interactions of a molecule or ion with its surrounding solvent. Desolvation removes or replaces these interactions, as when a group becomes buried.
- steric clash
- Nonbonded atoms closer than their van der Waals radii allow: very costly, and a sign of model error.
- Lennard-Jones potential
- The 12-6 formula force fields use for van der Waals energy, a steep repulsive wall plus a shallow attractive well.
- Coulomb's law
- The energy between two charges: proportional to their product, inversely proportional to distance and dielectric constant.
- van der Waals radius
- An atom's effective size in nonbonded contacts, about 1.7 Å for carbon. Atoms closer than the sum of their radii overlap.
1.5 Water and the hydrophobic effect
- hydrophobic
- Water-avoiding: nonpolar, mostly carbon and hydrogen, with nothing for water to hydrogen bond to.
- hydrophobic effect
- The solvent-dependent tendency of nonpolar groups to reduce their exposure to water. Its free-energy balance includes the surrounding water and varies with solute size, temperature, and geometry.
- free energy
- G = H − TS. At fixed temperature and pressure, a negative change in G favors a process; the rate also depends on barriers.
- enthalpy
- Heat content H; for molecules, mostly the energy of their bonds and interactions.
- entropy
- A measure of the number and probabilities of microscopic arrangements. For W equally probable arrangements, S = k ln W.
- hydrophilic
- Water-loving: polar or charged, forming hydrogen bonds or ionic interactions with water.
- hydrophobic core
- A folded protein's interior, packed mostly with hydrophobic side chains away from water.
1.6 Chirality and disulfide bonds
- chirality
- Handedness: a chiral molecule can't be superimposed on its mirror image, usually because a carbon carries four different groups.
- L-amino acid
- L is the standard handedness of chiral amino acids incorporated by ribosomes; D is its mirror form. D residues occur in specialized pathways and some modified peptides; glycine is achiral.
- achiral
- Identical to its mirror image. Glycine is the one achiral standard amino acid.
- CORN rule
- With the α-carbon's hydrogen toward you, an L-amino acid reads CO, R, N clockwise.
- disulfide bond
- A covalent S–S link, about 2.0 Å long, formed by oxidation of two cysteine thiols. It can constrain one chain or connect separate chains; suitable geometry and chemical conditions are required.
- oxidation
- Loss of electrons. Disulfide formation oxidizes two thiols; reduction reverses the process. The cytosol usually favors reduced thiols, while some internal compartments favor disulfides.
Lab 1 Interactions in real structures
- tautomer
- One of interconvertible chemical forms differing in proton positions and associated bonding, such as the two neutral histidine ring forms.
2.1 Amino acids and the peptide bond
- residue
- One monomer in a chain: an amino acid in a protein, or a nucleotide in DNA or RNA. Joining the chain costs each a water molecule, hence the name.
- backbone
- The repeating chain along a polymer: N–Cα–C (plus each C's oxygen) in a protein, sugar–phosphate in DNA and RNA. Side chains or bases hang off it.
- peptide bond
- The C–N bond from one residue's carboxyl carbon to the next residue's amino nitrogen. Partial double-bond character keeps it flat and unable to rotate.
- amino acid
- A molecule with an amino group, a carboxyl group, a hydrogen, and a side chain on one central carbon (Cα). Twenty kinds build proteins.
- primary structure
- A protein's amino-acid sequence, read from N- to C-terminus.
- side chain
- The part of an amino acid that varies, attached to Cα. It sets the residue's size, charge, and preference for water or oil.
- Cα
- The central carbon of an amino acid, bonded to the amino group, the carboxyl group, a hydrogen, and the side chain. Named CA in structure files.
- zwitterion
- A neutral molecule carrying both a positive and a negative charge, as a free amino acid does in water (–NH₃⁺ and –COO⁻).
- polypeptide
- A chain of amino acids joined by peptide bonds. A protein is one or more polypeptides, folded.
- condensation reaction
- Joining two molecules with the release of water. Peptide bonds form this way.
- N-terminus
- The start of a protein chain, with the free amino group. Sequences are written and numbered from here.
- C-terminus
- The end of a protein chain, carrying the free carboxyl group.
- Cβ
- The first side-chain carbon, bonded to Cα; every amino acid but glycine has one. Side-chain atoms are lettered β, γ, δ, ε, ζ outward.
- ω
- The torsion about the peptide bond: nearly always near 180° (trans). About 5% of bonds before a proline are cis, near 0°.
- trans
- Opposite sides: the two Cα atoms across a peptide bond (ω near 180°, the usual form), matching groups across a double bond, or the two glycosidic bonds of a base pair.
- cis
- Same side: the two Cα atoms across a peptide bond (ω near 0°, rare except before proline), matching groups across a double bond, or the two glycosidic bonds of a base pair.
- FASTA
- The standard text format for sequences: a header line starting with ">", then the residues on one or more lines.
2.2 Meet the twenty
- aliphatic
- An open chain of carbon and hydrogen with no ring (A, V, L, I).
- charged residue
- A residue whose side chain carries a full charge at pH 7: negative for D and E, positive for K and R. Histidine is charged only some of the time.
- polar residue
- A residue with a neutral side chain that can donate or accept hydrogen bonds (S, T, N, Q, and by some counts Y and C).
- kinase
- An enzyme that puts a phosphate on a protein, usually on Ser, Thr, or Tyr, switching the protein's behavior.
- guanidinium
- Arginine's side-chain tip, one carbon bonded to three nitrogens that share a positive charge.
- imidazole
- Histidine's five-membered ring with two nitrogens. Its pKa near 6 lets it switch between neutral and positive.
- hydrophobic residue
- A residue whose side chain is mostly carbon and hydrogen and can't hydrogen-bond to water (A, V, L, I, M, F, W). These tend to be buried.
2.3 Reading a sequence
- hydropathy
- A number per amino acid for how much its side chain prefers oil to water; positive is hydrophobic. Kyte and Doolittle's is the usual scale.
- transmembrane
- Crossing a membrane. A transmembrane helix is about 20 mostly hydrophobic residues, enough to span the bilayer's oily core.
- amphipathic
- Hydrophobic on one face and polar on the other, like a helix packed against a protein on one side and facing water on the other.
- helical wheel
- A view down a helix's axis that places each residue 100° from the previous one, showing which residues share a face.
- hydrophobic moment
- The vector sum of hydropathies, each pointing toward its residue on a helical wheel. Large when one face is oily and the other polar.
- low-complexity region
- A stretch built from few residue types, such as a run of glutamines; often disordered.
- signal peptide
- A hydrophobic stretch at a new protein's N-terminus that steers it to a membrane and is then cut off.
2.4 Burial and surfaces
- solvent-excluded surface
- The boundary of the space a rolling probe can't enter, hugging the atoms and closing over crevices too narrow for water; what viewers draw as "the surface".
- solvent-accessible surface
- The surface traced by the center of a probe sphere (1.4 Å, the size of a water molecule) rolled over a molecule's atoms.
- relative solvent accessibility
- A residue's accessible area over its maximum possible. Below about 0.2 the residue is usually called buried.
3.1 Degrees of freedom
- dihedral angle
- The signed angle, −180° to 180°, between the two planes of four consecutive atoms, viewed down the middle bond.
- φ
- The backbone torsion about the N–Cα bond, defined by C(i−1), N, Cα, C.
- ψ
- The backbone torsion about the Cα–C bond, defined by N, Cα, C, N(i+1).
- NeRF
- Natural extension reference frame: placing each next atom of a chain from the three before it, given a bond length, an angle, and a torsion.
- internal coordinates
- Bond lengths, bond angles, and torsions in place of x, y, z per atom.
- RMSD
- Root-mean-square deviation between matching atoms of two structures after the best rigid superposition. Under 1 Å is essentially the same structure.
3.2 The Ramachandran plot
- steric hindrance
- Atoms in each other's way. A conformation that pushes atoms closer than their sizes allow is forbidden.
- Ramachandran plot
- ψ plotted against φ for a protein's residues, showing which backbone conformations are allowed and which push atoms into each other.
- αR region
- The Ramachandran region near (φ, ψ) = (−63°, −43°), used by right-handed helices.
- polyproline II
- An extended, left-handed helical conformation near (−75°, +145°), common in loops and unfolded chains.
- αL region
- The small region near (+57°, +47°), a left-handed helical conformation used mostly by glycine.
- MolProbity
- The standard program for checking a structure's geometry, including Ramachandran outliers.
3.3 Helices, sheets, turns, and loops
- α-helix
- A right-handed spiral of the backbone with 3.6 residues and 5.4 Å per turn, held by a hydrogen bond from each C=O to the N–H four residues ahead.
- β-strand
- A nearly fully extended stretch of backbone. Strands pair side by side through hydrogen bonds into sheets.
- secondary structure
- The regular local shapes of a protein backbone, chiefly α-helices and β-strands, held by backbone hydrogen bonds.
- loop
- A stretch of chain with no regular structure: in a protein, backbone that is neither helix nor strand, often ordered and functional; in RNA, unpaired bases between or at the end of stems.
- 3₁₀ helix
- A tighter helix, 3 residues per turn with i → i+3 hydrogen bonds. Usually short, often at the end of an α-helix.
- π-helix
- A wider helix, about 4.3 residues per turn with i → i+5 hydrogen bonds; usually a one-turn bulge in an α-helix.
- β-sheet
- Two or more β-strands side by side, joined by backbone hydrogen bonds, with side chains alternating above and below.
- parallel sheet
- A sheet whose neighboring strands run in the same N→C direction.
- antiparallel sheet
- A sheet whose neighboring strands run in opposite N→C directions.
- DSSP
- The standard program that assigns secondary structure from backbone hydrogen bonds, in eight states (H, G, I, E, B, T, S, blank).
- β-hairpin
- Two antiparallel strands joined by a short turn.
- β-turn
- A four-residue reversal of the chain, closed by a hydrogen bond from the C=O of the first residue to the N–H of the fourth.
- Q3
- The fraction of residues whose three-state secondary-structure label (helix, strand, coil) a prediction gets right.
- contact map
- A residue-by-residue matrix marking pairs close in 3D, usually within 8 Å. Unchanged by rotating or moving the molecule.
3.4 Motifs, domains, and folds
- motif
- A small recurring pattern. In structure, a few secondary-structure elements such as a β-hairpin or helix-turn-helix; in sequence, a short conserved run such as a kinase's DFG.
- domain
- A compact part of a protein that usually folds on its own, sometimes built from pieces far apart in sequence.
- tertiary structure
- The 3D fold of one whole chain.
- quaternary structure
- The arrangement of several chains in a complex.
- protein fold
- A chain's overall 3D architecture: the arrangement and connection order of its helices and strands.
- topology
- The order, pairing, and direction of a protein's helices and strands.
- inhibitor
- A molecule that blocks an enzyme, often a drug occupying its active site.
- fold class
- The coarsest grouping of folds by secondary structure, as all-α, all-β, α/β (helices and parallel strands interleaved), or α+β (helices and mostly antiparallel strands kept apart).
- SCOP
- A database that classifies protein domains by hand into class, fold, superfamily, and family.
- CATH
- A database that classifies protein domains by Class, Architecture, Topology, and Homologous superfamily.
3.5 Beyond the soluble globule
- globular protein
- A water-soluble protein folded compactly, with a hydrophobic core and a polar surface.
- cofactor
- A non-protein molecule or metal ion a protein needs for its chemistry or structure, such as heme or zinc.
- fibrous protein
- A protein built as long fibers from repeated sequence, such as collagen or keratin's coiled coils.
- collagen
- The rope-like protein of skin and tendon: three chains in a triple helix, with glycine at every third position.
- coiled coil
- Two or more α-helices wound around each other, with a seven-residue repeat whose first and fourth positions are hydrophobic.
- hydroxyproline
- Proline hydroxylated after the chain is made. It stabilizes collagen's triple helix.
- intrinsically disordered
- Having no single stable structure on its own. Many such proteins and regions fold only when they bind a partner.
- zinc finger
- A small DNA-binding domain: a β-hairpin and a helix held together by a zinc ion bound to cysteines and histidines.
- glycosylation
- Attaching sugar chains to a protein, common on proteins that work outside cells.
- phosphorylation
- A kinase adding a phosphate to Ser, Thr, or Tyr: about two negative charges, often switching the protein on or off.
- MODRES
- A structure-file record naming a modified residue and the standard amino acid it came from.
- post-translational modification
- A chemical change to a protein after it is made, such as an added phosphate or sugar. The encoded sequence is unchanged.
3.6 Quaternary structure and hemoglobin
- interface
- The surface where two chains or molecules touch, measured by the accessible area each loses.
- subunit
- One part of a multi-part assembly: a chain in a multi-chain protein, or one of the ribosome's two RNA–protein halves.
- homomer
- A complex of identical chains. A heteromer has different chains, like hemoglobin's α₂β₂.
- tetramer
- A complex of four chains. A dimer has two.
- cooperativity
- Binding at one site changing the affinity of the others. It gives hemoglobin its S-shaped oxygen curve.
- 2,3-BPG
- A strongly negative small molecule in red blood cells that binds deoxyhemoglobin's central cavity and lowers its oxygen affinity.
- heme
- A flat ring holding an iron atom. In hemoglobin and myoglobin it binds the oxygen.
- myoglobin
- Muscle's single-chain oxygen-binding protein, with the same globin fold and heme pocket as one hemoglobin chain.
- Bohr effect
- Hemoglobin's oxygen affinity falling as pH falls, which releases oxygen in active tissue.
- T state
- Hemoglobin's two quaternary arrangements, T (tense, deoxy, low oxygen affinity) and R (relaxed, oxy, high affinity).
4.1 Energy, entropy, and free energy
- mole
- 6.022 × 10²³ of something, usually molecules. Quoting energies per mole scales them to laboratory amounts.
- second law of thermodynamics
- Total entropy never decreases. Local order, such as a folded protein, is paid for by heat released to the surroundings; ΔG < 0 is the law stated for the system alone.
- hydrolysis
- Splitting a bond by adding water across it, the reverse of condensation. It breaks proteins into amino acids and ATP into ADP.
- internal energy
- The energy E stored in a molecule's bonds and contacts. A force field computes it from coordinates and leaves out entropy.
- gas constant
- R = 1.987 × 10⁻³ kcal/(mol·K), Boltzmann's constant per mole (R = N_A k_B).
- ATP
- The cell's energy carrier. Hydrolyzing its last phosphate releases about 7.3 kcal/mol under standard conditions.
- melting temperature
- The temperature at which half a protein's molecules are unfolded, where ΔG of folding is zero.
- enthalpy–entropy compensation
- A gain in enthalpy (a tighter contact) offset by a loss of entropy (frozen motion), so the net free energy changes less than either part.
4.2 The Boltzmann distribution
- Metropolis Monte Carlo
- A sampler for the Boltzmann distribution: propose a random move, always accept it if the energy drops, and accept an uphill move with probability e^(−ΔE/kT).
- molecular dynamics
- Moving atoms by integrating Newton's equations with force-field forces at a set temperature. It samples the Boltzmann distribution over time.
- microstate
- One complete arrangement of a system with every atom placed; for a protein, one conformation.
- macrostate
- A set of microstates treated as one state, such as "folded" or "bound". Its free energy sums over its members.
- partition function
- The normalizer of the Boltzmann distribution, Z = Σ e^(−E/kT). Its log gives the free energy, F = −kT ln Z.
- equilibrium constant
- The ratio of two states' populations or concentrations at equilibrium, K = e^(−ΔG/RT). Each factor of ten is 1.36 kcal/mol at room temperature.
- Boltzmann distribution
- At equilibrium, the probability of a microstate with energy E is proportional to e^(−E/kT): a softmax over negative energies with temperature kT.
- ergodic hypothesis
- The assumption that a long enough simulation visits each region in proportion to its equilibrium probability, so time averages equal ensemble averages.
- detailed balance
- At equilibrium, moves from A to B happen as often as moves from B to A. Metropolis acceptance is built to satisfy it.
- simulated annealing
- Monte Carlo cooled slowly from a high temperature, crossing barriers early and settling into a low minimum late. An optimizer, not a sampler.
- sampling problem
- A simulation failing to visit all important states in the time available, usually trapped by barriers. Often a bigger limit than the energy function.
4.3 Binding and Kd
- dissociation constant
- For P + L ⇌ PL, Kd = [P][L]/[PL]: the free ligand concentration at which half the protein is bound. Smaller means tighter binding; ΔG° = RT ln Kd.
- binding free energy
- The free-energy change on binding, ΔG = RT ln Kd: about −8 kcal/mol at 1 µM and −12 kcal/mol at 1 nM.
- standard state
- The reference conditions (1 M, 1 atm) that make an equilibrium constant a pure number. The ° in ΔG° marks them.
- fraction bound
- The fraction θ of protein molecules with ligand bound, θ = [L]/(Kd + [L]) for a single site.
- Hill coefficient
- The exponent n in θ = pⁿ/(P50ⁿ + pⁿ), measuring how steep a binding curve is: 1 for independent sites, above 1 for positive cooperativity (about 2.8 for hemoglobin).
- partial pressure
- One gas's share of a mixture's pressure, used as its concentration; in mm Hg or torr, which are the same unit.
- scoring function
- A fast estimate of binding free energy used to rank docked poses and compounds. Typical errors are about 2 kcal/mol, a factor of 30 in Kd.
- knowledge-based potential
- An energy term from how often contacts occur in known structures, by inverting the Boltzmann relation: E ≈ −RT ln(frequency).
- IC50
- The inhibitor concentration that halves an activity. Related to Kd but dependent on the assay.
- P50
- The oxygen pressure at which a carrier is half saturated, the gas-phase equivalent of Kd: about 26 torr for hemoglobin and 2.8 torr for myoglobin.
4.4 Folding: Anfinsen, Levinthal, and funnels
- denaturation
- Loss of a protein's folded structure, by heat or chemicals such as urea, with the chain left intact.
- urea
- A small polar molecule that at high concentration unfolds proteins by competing for their hydrogen bonds and solvating their interiors.
- thermodynamic hypothesis
- Anfinsen's conclusion from refolding ribonuclease: under physiological conditions the native structure is the lowest-free-energy state, and the sequence determines it.
- native state
- A protein's folded, functional structure under normal conditions.
- reducing agent
- A chemical that donates electrons, here turning disulfide bonds back into free –SH groups. β-Mercaptoethanol is the classic one.
- chaperone
- A protein that helps others fold by keeping partly folded chains from sticking together and giving misfolded ones another try. It doesn't encode the structure.
- amyloid
- A stack of β-strands formed when many copies of a protein aggregate. Often more stable than the native fold; a hallmark of Alzheimer's and prion diseases.
- marginal stability
- The small net free energy of folding, typically 5–15 kcal/mol, a difference of much larger enthalpy and entropy terms. One mutation can erase it.
- two-state folding
- Each molecule is either essentially folded or unfolded, with intermediates too rare to detect. Typical of small proteins.
- Levinthal's paradox
- A random search of a chain's conformations would take longer than the age of the universe, yet proteins fold in milliseconds to seconds, so folding isn't a random search.
- folding funnel
- Folding pictured as descent on a funnel-shaped energy landscape, where most downhill moves lead toward the native state. Bumps on the walls are kinetic traps.
- kinetic trap
- A low-energy non-native state, separated from the native one by a barrier, where a folding chain or a simulation gets stuck.
- energy minimization
- Moving atoms downhill on the energy function to the nearest local minimum, like gradient descent. It removes clashes but never crosses barriers.
- solvation term
- An energy-function term that stands in for unmodeled water, including the hydrophobic effect.
- ΔΔG
- The change in a free energy of folding or binding caused by a mutation. Sign conventions vary, so check which one is used.
Lab 4 A lattice protein
- density of states
- The number of conformations at each energy, g(E). The partition function is Σ g(E) e^(−E/kT).
- HP model
- A lattice folding model with hydrophobic (H) and polar (P) residues, scoring −1 per H–H contact. Small enough to enumerate exactly.
- heat capacity
- Energy change per degree, C = Var(E)/kT². It peaks at a melting transition, sharply for a two-state protein.
5.1 Mutation, divergence, and homology
- homolog
- A protein or gene that shares a common ancestor with another. Homology is yes-or-no; similarity is measured.
- homology modeling
- Predicting a structure by adapting the known structure of a homolog (the template).
- evolution
- Change in the genetic makeup of populations over time, driven by mutation, natural selection, and genetic drift.
- mutation
- A change in DNA sequence: a substitution, insertion, or deletion. Most are neutral or harmful; a few help.
- genome
- An organism's complete DNA.
- natural selection
- Variants that help survival and reproduction becoming more common, and harmful ones disappearing.
- allele
- One version of a gene among those present in a population.
- species
- Organisms that can interbreed. Once two populations stop interbreeding, their genomes diverge independently.
- genetic drift
- Random change in an allele's frequency between generations, regardless of its effect. It spreads or removes neutral variants.
- gene duplication
- An accidental copy of a gene within a genome. The copies then diverge, often in function; the globins arose this way.
- phylogenetic tree
- A tree of descent: sequences at the leaves, ancestors at internal nodes, branch lengths showing amount of change.
- clade
- An ancestor together with all of its descendants: a subtree of a phylogenetic tree.
- synonymous substitution
- A DNA change that keeps the encoded amino acid, usually at a codon's third position. Selection barely sees it.
- missense mutation
- A DNA change that replaces one amino acid with another, such as the sickle-cell mutation in β-globin.
- frame-shift mutation
- An insertion or deletion not a multiple of three bases, which shifts the reading frame and scrambles every later codon.
- wild type
- The common, normal version of a gene or protein, the reference for describing mutants.
- indel
- An insertion or deletion; a gap in an alignment.
- neutral theory
- Kimura's proposal that most molecular variation is selectively neutral and spreads by drift.
- purifying selection
- Selection removing harmful changes, which keeps important positions, such as buried and active-site residues, conserved.
- ortholog
- Homologs split by a speciation: the same gene in two species, like human and whale myoglobin.
- paralog
- Homologs split by a gene duplication, like myoglobin and the hemoglobin chains.
- analog
- Proteins alike in shape or job without a common ancestor: evolution found the same solution twice.
- sequence identity
- The fraction of aligned positions where two sequences have the same residue.
- twilight zone
- About 20–40% sequence identity, where homologs and chance matches overlap. Below about 20% (the midnight zone) homology is hard to detect from sequence.
- protein language model
- A neural network trained to fill in masked residues across hundreds of millions of protein sequences. It learns family constraints, including coevolution, implicitly.
5.2 Alignment and substitution matrices
- pairwise alignment
- A residue-by-residue correspondence between two sequences, with gaps, chosen to maximize a score. Found by dynamic programming.
- substitution matrix
- Scores for aligning each pair of amino acids: the log-odds of how often relatives pair them versus chance.
- BLAST
- The standard fast tool for searching sequence databases for similar sequences.
- gap penalty
- The cost of a gap. Affine penalties charge more to open a gap than to extend it, since one long indel is likelier than several short ones.
- Needleman–Wunsch
- Dynamic programming for the best global alignment, using every residue of both sequences.
- Smith–Waterman
- Dynamic programming for the best local alignment, the highest-scoring pair of segments; a floor of zero cuts off poor stretches.
- BLOSUM62
- The standard substitution matrix, from gap-free blocks of related proteins clustered at 62% identity, in half-bit log-odds units.
- log-odds
- The log of a probability ratio, here how often a pair appears in true alignments over how often by chance. Summed, they score a likelihood ratio.
- E-value
- The number of hits scoring this well expected by chance in a search this size. Near 1 or above is consistent with chance.
5.3 Families, alignments, and profiles
- multiple sequence alignment
- Many related sequences stacked so that each column holds residues descended from the same ancestral position.
- Pfam
- A database of over 20,000 protein families, each with a curated seed alignment, a profile HMM, and a full alignment of members from UniProt.
- conservation
- How little a position varies across a family, measured by entropy or information content. Conserved positions are usually core or functional.
- substrate
- The molecule an enzyme acts on.
- information content
- log₂ 20 minus a column's Shannon entropy: high where the family allows few residues. A sequence logo draws it as stack height.
- sequence logo
- A profile drawn as letter stacks: each column's height is its information content, and each letter's height its frequency.
- sequence weighting
- Down-weighting near-duplicate sequences (often by 1 over the number of neighbors above 80% identity) so clusters don't dominate. The weights sum to the effective number of sequences.
- pseudocount
- A few pretend observations added to alignment counts so unseen residues get small probabilities, not zero.
- position-specific scoring matrix
- A profile turned into scores: residue a at position i scores log(q_i(a)/p_a), a different log-odds column at every position.
- A profile with match, insert, and delete states and transition probabilities between them. Aligning to it uses dynamic programming (Viterbi).
- sequence profile
- Per-column amino-acid probabilities computed from a multiple sequence alignment.
5.4 Coevolution and contacts
- coevolution
- Correlated change at two positions across a family, often because the residues touch and one change compensates for the other. The signal behind contact prediction and AlphaFold's use of alignments.
- mutual information
- How much one alignment column tells you about another; zero for independent columns. Shared ancestry and indirect chains inflate it.
- average product correction
- Subtracting from each pair's mutual information the part explained by the two columns' average mutual information, which removes much of the phylogenetic background.
- Potts model
- A Boltzmann distribution over sequences with a field for each position and a coupling for each pair, fit to reproduce an alignment's single and pair frequencies.
- direct coupling analysis
- Ranking residue pairs by the couplings of a fitted Potts model, which separates direct contacts from correlations relayed through other residues.
- contact prediction
- Predicting which residue pairs touch in the folded protein, from coevolution or a neural network, as an input to 3D prediction.
6.1 Models, not photographs
- X-ray crystallography
- Determining a structure from the diffraction spots of X-rays passed through a crystal, by computing the electron density.
- crystal
- A solid in which one small box of molecules, the unit cell, repeats in three dimensions.
- NMR
- Nuclear magnetic resonance: nuclei in a strong magnetic field absorb radio waves at frequencies that reveal their chemical surroundings and nearby atoms.
- cryo-EM
- Imaging frozen molecules in an electron microscope and combining many noisy 2D particle images into a 3D map.
- diffraction
- Waves scattered by a repeating object reinforce in a few directions and cancel elsewhere, so a crystal turns X-rays into a pattern of sharp spots.
- electron density
- Electrons per cubic ångström at each point of the crystal, computed from the diffraction data. Atoms are its peaks.
- unit cell
- The repeating box of a crystal; stacked copies of it make the whole crystal.
- reflection
- One spot of a diffraction pattern, corresponding to one Fourier coefficient of the electron density.
- phase problem
- A detector records each reflection's intensity but not its phase, and the phases decide what the density map looks like.
- molecular replacement
- Getting starting phases by placing a known similar structure (or a prediction) in the new crystal by rotation and translation searches.
- refinement
- Adjusting a model's coordinates, B factors, and occupancies to fit the diffraction data, with ideal bond geometry as a restraint.
- restraint
- A bound, such as a maximum distance between two atoms, that a model must satisfy during calculation or refinement.
- spin
- The property that makes some nuclei, including hydrogen's, behave like tiny magnets.
- chemical shift
- The small shift in a nucleus's NMR frequency caused by its chemical surroundings, which gives each hydrogen its own peak.
- NOE
- Nuclear Overhauser effect: magnetization transfer between nuclei close in space, falling off as 1/r⁶, so an NOE puts two hydrogens within about 5–6 Å.
- ensemble
- A set of structures describing one molecule: the conformations it moves through, or the 10 to 20 models deposited together for an NMR entry, each consistent with the data.
- distance geometry
- Finding coordinates for points from known or bounded pairwise distances; NP-hard when only some distances are known.
- vitrified ice
- Water frozen so fast that it becomes a glass instead of ice crystals, keeping molecules in their solution shapes.
- particle (cryo-EM)
- One copy of the molecule caught in the ice, in some orientation, imaged as a noisy 2D projection.
- homology model
- A protein model built from a related protein's experimental structure, with no measurement of the protein itself.
- pLDDT
- AlphaFold's per-residue confidence, 0 to 100, stored in the B-factor column. High is good, the opposite sense of a B factor.
- PAE
- AlphaFold's expected error in one residue's position when the model is aligned on another. It shows how confidently domains are placed relative to each other.
- local resolution
- The resolution of one region of a cryo-EM map, which can be much worse than the map's overall figure.
- structure factor
- The complex Fourier coefficient F(h) behind one reflection: an amplitude, which is measured, and a phase, which is not.
6.2 Reading the fine print
- resolution
- The finest spacing of detail a structure determination resolved, in ångströms. Smaller is better; side chains are clear near 2 Å.
- R-free
- The R-factor on 5–10% of reflections left out of refinement: a validation error that exposes overfitting.
- occupancy
- The fraction of molecules in the crystal with an atom at the given position; alternate conformations share it.
- B factor
- The width of an atom's positional smear in the model, in Ų (B = 8π²⟨u²⟩). It mixes vibration, static disorder, and model error.
- Bragg's law
- λ = 2d sin θ, relating the X-ray wavelength, the spacing d of planes in the crystal, and the angle at which they diffract.
- R-factor
- The relative disagreement between measured and model-computed structure-factor amplitudes. About 0.20 is typical for proteins.
- space group
- The symmetry operations that relate copies of the molecule within a crystal.
- static disorder
- Molecules of a crystal frozen in slightly different positions. It raises B factors just as motion does.
- alternate conformation
- One of two or more positions a group takes in different molecules of the crystal, labeled A, B, … in the altloc field.
- validation report
- The PDB's per-entry assessment of geometry, clashes, Ramachandran and side-chain outliers, and fit to the data, ranked against similar entries.
- clashscore
- Serious steric overlaps per 1,000 atoms, counted after hydrogens are added; part of the PDB validation report.
6.3 PDB and mmCIF files
- PDB format
- The original fixed-column text format for structures, with fields defined by column ranges, not spaces.
- ATOM record
- A legacy PDB line for one atom of a standard polymer residue, with its coordinates, occupancy, and B factor.
- chain ID
- The label for which polymer chain an atom belongs to. Author (auth_asym_id) and archive (label_asym_id) versions can differ.
- insertion code
- A letter after a residue number (60A, 60B) for residues inserted without renumbering the rest.
- mmCIF
- The PDB's master format: named tables (categories) such as _atom_site, with no size limits.
- SEQRES
- Records of the full sequence of each polymer studied, including residues with no coordinates.
- The residue number the depositors chose; it can start anywhere, skip, or carry an insertion code.
- UniProt
- The main protein-sequence database. Its positions are a protein's reference numbering.
- expression tag
- Extra residues added to make a protein easy to purify, often left in the deposited sequence.
- asymmetric unit
- The smallest part of a crystal from which symmetry rebuilds the rest. An entry's coordinates usually cover only this.
- crystal contact
- A place where neighboring molecules in a crystal touch. It can hold loops in place and looks like an interface without being one.
- biological assembly
- The functional complex, rebuilt from the asymmetric unit with the symmetry operations stored in the entry.
6.4 Looking at structures
- space-filling
- Each atom drawn as a sphere of its van der Waals radius, showing shape and hiding the interior.
- ball and stick
- Atoms as small spheres joined by cylinders for bonds; good for geometry on a few residues.
- cartoon
- A protein drawing with helices as spirals or cylinders, strands as arrows, and loops as tubes; side chains are hidden.
- Cα trace
- A line joining consecutive Cα atoms, showing only the chain's path.
- CONECT
- Optional records listing bonds; without them, viewers infer bonds from distances.
- metal coordination
- Bonds between a metal ion and atoms that supply both shared electrons, such as the histidine nitrogen holding heme iron.
- χ1
- The first side-chain torsion angle, about the Cα–Cβ bond.
7.1 The double helix
- transcription factor
- A protein that switches a gene on or off by binding DNA next to it.
- ribose
- RNA's pentose, with a hydroxyl (2′-OH) on its 2′ carbon.
- deoxyribose
- DNA's pentose: ribose with the 2′ hydroxyl replaced by hydrogen.
- pentose
- A five-carbon sugar; ribose in RNA and deoxyribose in DNA.
- nucleoside
- A base joined to a sugar, without the phosphate.
- glycosidic bond
- The bond from a sugar's C1′ to its base: to N9 of a purine or N1 of a pyrimidine.
- anti
- The usual glycosidic orientation, with the base's pairing edge pointing away from the sugar.
- antiparallel
- Running in opposite directions. A duplex pairs a 5′→3′ strand against a 3′→5′ one.
- phosphodiester bond
- The link between nucleotides: a phosphate joining one sugar's 3′ carbon to the next sugar's 5′ carbon.
- major groove
- B-DNA's wider groove, where each base pair shows a distinct pattern that proteins read.
- minor groove
- B-DNA's narrower groove, where base pairs look nearly alike; small drugs often bind there.
- B-form
- The right-handed DNA helix of cells: 10–10.5 pairs per turn, 3.4 Å rise, pairs on the axis, a wide major and a narrow minor groove.
- Watson–Crick pair
- The canonical pairs A–T (A–U in RNA), with two hydrogen bonds, and G–C, with three.
- base stacking
- Flat bases packed face to face 3.4 Å apart, the main source of a duplex's stability.
- nearest-neighbor model
- A duplex's stability as a sum over neighboring base-pair steps, each with a measured free energy.
- sugar pucker
- Which ring atom of the sugar sits out of plane, and on which side: C2′-endo in B-DNA, C3′-endo in RNA.
- pseudorotation phase
- The angle P summarizing a five-membered ring's pucker from its five ring torsions. North (near 18°) is C3′-endo, south (near 160°) C2′-endo.
- A-form
- The right-handed helix of RNA duplexes: 11 pairs per turn, 2.6 Å rise, C3′-endo sugars, pairs off the axis.
- syn
- The glycosidic orientation with the base turned over the sugar; seen in Z-DNA's guanines.
- Z-DNA
- A left-handed DNA helix formed by alternating C and G, with syn guanines and a zigzag backbone.
- intercalation
- A flat molecule slipping between two base pairs, lengthening and unwinding the helix.
- covalent adduct
- A molecule chemically bonded to DNA, as cisplatin is to guanines.
- 5′ end
- The ends of a nucleic-acid strand, named for the sugar carbon left free. Sequences are written 5′ to 3′.
- reverse complement
- A sequence's partner strand read 5′ to 3′: complement each base and reverse the order.
7.2 How RNA folds
- stem
- A run of consecutive base pairs in an RNA, forming a short A-form helix.
- hairpin loop
- Unpaired bases closed by one pair, capping the end of a stem.
- bulge
- Unpaired bases on one side of a stem, between two paired stretches.
- internal loop
- Unpaired bases on both sides between two stems.
- multiloop
- A loop from which three or more stems leave.
- pseudoknot
- Pairs that cross other pairs, so two stems interleave. They break the nesting that folding algorithms assume.
- dot-bracket
- An RNA structure as a string: '.' for unpaired, '(' and ')' for the two ends of each pair.
- secondary structure (RNA)
- The set of base pairs an RNA forms with itself, drawn as stems and loops.
- Nussinov algorithm
- Dynamic programming that finds the nested structure with the most base pairs in O(n³) time.
- Zuker algorithm
- The same recursion as Nussinov's, minimizing nearest-neighbor free energy instead of counting pairs.
- wobble pair
- The G·U pair, with two hydrogen bonds; common in RNA stems and at the third codon position.
- anticodon
- The three bases of a tRNA that pair with a codon of the mRNA.
- Hoogsteen edge
- One of a purine's three pairing edges, with the Watson–Crick and sugar edges.
- noncanonical pair
- Any base pair other than Watson–Crick and G·U, classed by which edges meet and whether the glycosidic bonds are cis or trans.
- tertiary contact
- A contact between distant parts of a folded RNA that holds its 3D shape, often noncanonical and crossing the secondary structure.
- ribozyme
- An RNA that catalyzes a chemical reaction.
- riboswitch
- A stretch at the start of an mRNA that binds a small molecule and changes shape, turning its gene's protein production up or down.
- rRNA
- The ribosome's RNA, which forms most of its structure and its catalytic site.
Lab 7 Reading tRNA
- modified nucleotide
- A nucleotide chemically altered after the RNA is made, such as pseudouridine (Ψ). Structure files give each its own residue name.
- 2,3-BPG
- A strongly negative small molecule in red blood cells that binds deoxyhemoglobin's central cavity and lowers its oxygen affinity. Lesson 3.6
- 3₁₀ helix
- A tighter helix, 3 residues per turn with i → i+3 hydrogen bonds. Usually short, often at the end of an α-helix. Lesson 3.3
- 5′ end
- The ends of a nucleic-acid strand, named for the sugar carbon left free. Sequences are written 5′ to 3′. Lesson 7.1
- A-form
- The right-handed helix of RNA duplexes: 11 pairs per turn, 2.6 Å rise, C3′-endo sugars, pairs off the axis. Lesson 7.1
- acceptor
- In a hydrogen bond, the N or O atom the hydrogen points at, which offers a lone pair of electrons. Lesson 1.4
- achiral
- Identical to its mirror image. Glycine is the one achiral standard amino acid. Lesson 1.6
- acid (proton donor)
- Gives up a proton (H⁺). A carboxyl group is neutral with its proton and negative without it. Lesson 1.3
- aliphatic
- An open chain of carbon and hydrogen with no ring (A, V, L, I). Lesson 2.2
- allele
- One version of a gene among those present in a population. Lesson 5.1
- alternate conformation
- One of two or more positions a group takes in different molecules of the crystal, labeled A, B, … in the altloc field. Lesson 6.2
- amide
- –C(=O)–NH–, a carbonyl bonded to a nitrogen. The peptide bond is an amide. Lesson 0.4
- amino acid
- A molecule with an amino group, a carboxyl group, a hydrogen, and a side chain on one central carbon (Cα). Twenty kinds build proteins. Lesson 2.1
- amino group
- –NH₂. At pH 7 it usually gains an H to become ammonium, –NH₃⁺. Lesson 0.4
- amphipathic
- Hydrophobic on one face and polar on the other, like a helix packed against a protein on one side and facing water on the other. Lesson 2.3
- amyloid
- A stack of β-strands formed when many copies of a protein aggregate. Often more stable than the native fold; a hallmark of Alzheimer's and prion diseases. Lesson 4.4
- analog
- Proteins alike in shape or job without a common ancestor: evolution found the same solution twice. Lesson 5.1
- ångström
- 10⁻¹⁰ m, or 0.1 nm. Bond lengths are 1–2 Å, which is why structure files measure distance in ångströms. Lesson 0.1
- anion
- A negatively charged ion. Lesson 1.1
- anti
- The usual glycosidic orientation, with the base's pairing edge pointing away from the sugar. Lesson 7.1
- antibody
- An immune-system protein that binds a foreign molecule tightly and marks it for destruction. Lesson 0.1
- anticodon
- The three bases of a tRNA that pair with a codon of the mRNA. Lesson 7.2
- antiparallel
- Running in opposite directions. A duplex pairs a 5′→3′ strand against a 3′→5′ one. Lesson 7.1
- antiparallel sheet
- A sheet whose neighboring strands run in opposite N→C directions. Lesson 3.3
- aromatic
- A flat ring, like benzene, whose double-bond electrons are shared around it; drawn with alternating double bonds or a circle. Lesson 0.4
- asymmetric unit
- The smallest part of a crystal from which symmetry rebuilds the rest. An entry's coordinates usually cover only this. Lesson 6.3
- ATOM record
- A legacy PDB line for one atom of a standard polymer residue, with its coordinates, occupancy, and B factor. Lesson 6.3
- atomic number
- The number of protons in the nucleus, which sets the element. Lesson 1.1
- ATP
- The cell's energy carrier. Hydrolyzing its last phosphate releases about 7.3 kcal/mol under standard conditions. Lesson 4.1
- author residue number
- The residue number the depositors chose; it can start anywhere, skip, or carry an insertion code. Lesson 6.3
- average product correction
- Subtracting from each pair's mutual information the part explained by the two columns' average mutual information, which removes much of the phylogenetic background. Lesson 5.4
- B factor
- The width of an atom's positional smear in the model, in Ų (B = 8π²⟨u²⟩). It mixes vibration, static disorder, and model error. Lesson 6.2
- B-form
- The right-handed DNA helix of cells: 10–10.5 pairs per turn, 3.4 Å rise, pairs on the axis, a wide major and a narrow minor groove. Lesson 7.1
- backbone
- The repeating chain along a polymer: N–Cα–C (plus each C's oxygen) in a protein, sugar–phosphate in DNA and RNA. Side chains or bases hang off it. Lesson 2.1
- ball and stick
- Atoms as small spheres joined by cylinders for bonds; good for geometry on a few residues. Lesson 6.4
- base (of a nucleotide)
- The variable ring-shaped part of a nucleotide (A, C, G, T, or U), which carries the information. Bases on opposite strands pair by hydrogen bonds. Lesson 0.2
- base (proton acceptor)
- Takes up a proton (H⁺). An amino group is neutral without it and positive with it. Lesson 1.3
- base pair
- Two bases on opposite strands joined by hydrogen bonds; also the unit of length for double-stranded DNA. Lesson 0.2
- base stacking
- Flat bases packed face to face 3.4 Å apart, the main source of a duplex's stability. Lesson 7.1
- binding free energy
- The free-energy change on binding, ΔG = RT ln Kd: about −8 kcal/mol at 1 µM and −12 kcal/mol at 1 nM. Lesson 4.3
- biological assembly
- The functional complex, rebuilt from the asymmetric unit with the symmetry operations stored in the entry. Lesson 6.3
- BLAST
- The standard fast tool for searching sequence databases for similar sequences. Lesson 5.2
- BLOSUM62
- The standard substitution matrix, from gap-free blocks of related proteins clustered at 62% identity, in half-bit log-odds units. Lesson 5.2
- Bohr effect
- Hemoglobin's oxygen affinity falling as pH falls, which releases oxygen in active tissue. Lesson 3.6
- Boltzmann distribution
- At equilibrium, the probability of a microstate with energy E is proportional to e^(−E/kT): a softmax over negative energies with temperature kT. Lesson 4.2
- bond length
- The distance between two bonded atoms' centers; nearly constant for a given pair and bond type (C–C, 1.53 Å). Lesson 1.1
- Bragg's law
- λ = 2d sin θ, relating the X-ray wavelength, the spacing d of planes in the crystal, and the angle at which they diffract. Lesson 6.2
- bulge
- Unpaired bases on one side of a stem, between two paired stretches. Lesson 7.2
- C-terminus
- The end of a protein chain, carrying the free carboxyl group. Lesson 2.1
- carbonyl
- C=O. Every peptide bond has one. Lesson 0.4
- carboxyl
- –COOH. At pH 7 it loses the H to become carboxylate, –COO⁻. Lesson 0.4
- cartoon
- A protein drawing with helices as spirals or cylinders, strands as arrows, and loops as tubes; side chains are hidden. Lesson 6.4
- catalyst
- Something that speeds up a reaction and comes out unchanged. Lesson 0.1
- CATH
- A database that classifies protein domains by Class, Architecture, Topology, and Homologous superfamily. Lesson 3.4
- cation
- A positively charged ion. Lesson 1.1
- cell membrane
- The lipid bilayer and its proteins that enclose a cell and control what enters and leaves. Lesson 0.1
- cell nucleus
- The organelle that holds most of a eukaryotic cell's DNA. Lesson 0.1
- central dogma
- Information flows from DNA to RNA (transcription) and from RNA to protein (translation). Lesson 0.2
- chain ID
- The label for which polymer chain an atom belongs to. Author (auth_asym_id) and archive (label_asym_id) versions can differ. Lesson 6.3
- chaperone
- A protein that helps others fold by keeping partly folded chains from sticking together and giving misfolded ones another try. It doesn't encode the structure. Lesson 4.4
- charged residue
- A residue whose side chain carries a full charge at pH 7: negative for D and E, positive for K and R. Histidine is charged only some of the time. Lesson 2.2
- chemical shift
- The small shift in a nucleus's NMR frequency caused by its chemical surroundings, which gives each hydrogen its own peak. Lesson 6.1
- chirality
- Handedness: a chiral molecule can't be superimposed on its mirror image, usually because a carbon carries four different groups. Lesson 1.6
- cis
- Same side: the two Cα atoms across a peptide bond (ω near 0°, rare except before proline), matching groups across a double bond, or the two glycosidic bonds of a base pair. Lesson 2.1
- clade
- An ancestor together with all of its descendants: a subtree of a phylogenetic tree. Lesson 5.1
- clashscore
- Serious steric overlaps per 1,000 atoms, counted after hydrogens are added; part of the PDB validation report. Lesson 6.2
- coding strand
- The DNA strand whose sequence matches the RNA, with T for U. Lesson 0.2
- codon
- Three consecutive bases on an mRNA that specify one amino acid or a stop. Lesson 0.2
- coevolution
- Correlated change at two positions across a family, often because the residues touch and one change compensates for the other. The signal behind contact prediction and AlphaFold's use of alignments. Lesson 5.4
- cofactor
- A non-protein molecule or metal ion a protein needs for its chemistry or structure, such as heme or zinc. Lesson 3.5
- coiled coil
- Two or more α-helices wound around each other, with a seven-residue repeat whose first and fourth positions are hydrophobic. Lesson 3.5
- collagen
- The rope-like protein of skin and tendon: three chains in a triple helix, with glycine at every third position. Lesson 3.5
- complementary
- Paired by Watson–Crick rules: A with T (or U), G with C. Lesson 0.2
- condensation reaction
- Joining two molecules with the release of water. Peptide bonds form this way. Lesson 2.1
- condensed formula
- Each carbon written with its hydrogens, as in CH₃CH₂OH. Lesson 0.4
- CONECT
- Optional records listing bonds; without them, viewers infer bonds from distances. Lesson 6.4
- conformation
- One 3D arrangement of a molecule, reachable from others by rotating about single bonds. Lesson 1.1
- conservation
- How little a position varies across a family, measured by entropy or information content. Conserved positions are usually core or functional. Lesson 5.3
- contact map
- A residue-by-residue matrix marking pairs close in 3D, usually within 8 Å. Unchanged by rotating or moving the molecule. Lesson 3.3
- contact prediction
- Predicting which residue pairs touch in the folded protein, from coevolution or a neural network, as an input to 3D prediction. Lesson 5.4
- cooperativity
- Binding at one site changing the affinity of the others. It gives hemoglobin its S-shaped oxygen curve. Lesson 3.6
- CORN rule
- With the α-carbon's hydrogen toward you, an L-amino acid reads CO, R, N clockwise. Lesson 1.6
- Coulomb's law
- The energy between two charges: proportional to their product, inversely proportional to distance and dielectric constant. Lesson 1.4
- covalent adduct
- A molecule chemically bonded to DNA, as cisplatin is to guanines. Lesson 7.1
- covalent bond
- A chemical bond involving shared electron density. Ordinary covalent bonds remain intact during protein folding; changing them requires a chemical reaction. Lesson 1.1
- covalent radius
- An atom's share of a typical bond length. Atoms closer than the sum of their covalent radii, plus a tolerance, are taken as bonded. Lesson 1.1
- cryo-EM
- Imaging frozen molecules in an electron microscope and combining many noisy 2D particle images into a 3D map. Lesson 6.1
- crystal
- A solid in which one small box of molecules, the unit cell, repeats in three dimensions. Lesson 6.1
- crystal contact
- A place where neighboring molecules in a crystal touch. It can hold loops in place and looks like an interface without being one. Lesson 6.3
- crystal structure
- A structure determined by X-ray diffraction from a crystal. Lesson 0.3
- cytoplasm
- Everything inside the cell membrane (in a eukaryote, outside the nucleus). Lesson 0.1
- cytosol
- The watery part of the cytoplasm, crowded with proteins, salts, and small molecules. It is reducing, so disulfide bonds are rare there. Lesson 0.1
- Cα
- The central carbon of an amino acid, bonded to the amino group, the carboxyl group, a hydrogen, and the side chain. Named CA in structure files. Lesson 2.1
- Cα trace
- A line joining consecutive Cα atoms, showing only the chain's path. Lesson 6.4
- Cβ
- The first side-chain carbon, bonded to Cα; every amino acid but glycine has one. Side-chain atoms are lettered β, γ, δ, ε, ζ outward. Lesson 2.1
- dalton
- The unit of molecular mass, one-twelfth of a carbon-12 atom (about one hydrogen atom). A molecule of M daltons weighs M grams per mole. Lesson 0.1
- denaturation
- Loss of a protein's folded structure, by heat or chemicals such as urea, with the chain left intact. Lesson 4.4
- density of states
- The number of conformations at each energy, g(E). The partition function is Σ g(E) e^(−E/kT). Lab 4
- deoxyribose
- DNA's pentose: ribose with the 2′ hydroxyl replaced by hydrogen. Lesson 7.1
- detailed balance
- At equilibrium, moves from A to B happen as often as moves from B to A. Metropolis acceptance is built to satisfy it. Lesson 4.2
- dielectric constant
- A continuum measure of electrical polarization, about 80 for water near room temperature. Low values such as 2–4 are sometimes used for protein interiors, whose effective response depends on the model. Lesson 1.4
- diffraction
- Waves scattered by a repeating object reinforce in a few directions and cancel elsewhere, so a crystal turns X-rays into a pattern of sharp spots. Lesson 6.1
- dihedral angle
- The signed angle, −180° to 180°, between the two planes of four consecutive atoms, viewed down the middle bond. Lesson 3.1
- dipole moment
- The vector sum of charges times positions: how far apart a molecule's positive and negative ends are. Lesson 1.2
- direct coupling analysis
- Ranking residue pairs by the couplings of a fitted Potts model, which separates direct contacts from correlations relayed through other residues. Lesson 5.4
- dissociation constant
- For P + L ⇌ PL, Kd = [P][L]/[PL]: the free ligand concentration at which half the protein is bound. Smaller means tighter binding; ΔG° = RT ln Kd. Lesson 4.3
- distance geometry
- Finding coordinates for points from known or bounded pairwise distances; NP-hard when only some distances are known. Lesson 6.1
- disulfide bond
- A covalent S–S link, about 2.0 Å long, formed by oxidation of two cysteine thiols. It can constrain one chain or connect separate chains; suitable geometry and chemical conditions are required. Lesson 1.6
- DNA
- Deoxyribonucleic acid: a double-stranded nucleotide polymer that stores genetic information as a string over A, C, G, and T. Lesson 0.2
- domain
- A compact part of a protein that usually folds on its own, sometimes built from pieces far apart in sequence. Lesson 3.4
- donor
- In a hydrogen bond, the N or O atom that holds the hydrogen. Lesson 1.4
- dot-bracket
- An RNA structure as a string: '.' for unpaired, '(' and ')' for the two ends of each pair. Lesson 7.2
- double bond
- A bond represented by two shared electron pairs. A C=C bond is shorter than a C–C single bond and has a much larger rotational barrier. Lesson 1.1
- DSSP
- The standard program that assigns secondary structure from backbone hydrogen bonds, in eight states (H, G, I, E, B, T, S, blank). Lesson 3.3
- E-value
- The number of hits scoring this well expected by chance in a search this size. Near 1 or above is consistent with chance. Lesson 5.2
- electron density
- Electrons per cubic ångström at each point of the crystal, computed from the diffraction data. Atoms are its peaks. Lesson 6.1
- electronegativity
- How strongly an atom pulls shared electrons. O and N pull much harder than C and H. Lesson 1.2
- energy minimization
- Moving atoms downhill on the energy function to the nearest local minimum, like gradient descent. It removes clashes but never crosses barriers. Lesson 4.4
- ensemble
- A set of structures describing one molecule: the conformations it moves through, or the 10 to 20 models deposited together for an NMR entry, each consistent with the data. Lesson 6.1
- enthalpy
- Heat content H; for molecules, mostly the energy of their bonds and interactions. Lesson 1.5
- enthalpy–entropy compensation
- A gain in enthalpy (a tighter contact) offset by a loss of entropy (frozen motion), so the net free energy changes less than either part. Lesson 4.1
- entropy
- A measure of the number and probabilities of microscopic arrangements. For W equally probable arrangements, S = k ln W. Lesson 1.5
- enzyme
- A protein (occasionally an RNA) that catalyzes a particular reaction. Lesson 0.1
- equilibrium constant
- The ratio of two states' populations or concentrations at equilibrium, K = e^(−ΔG/RT). Each factor of ten is 1.36 kcal/mol at room temperature. Lesson 4.2
- equivariant
- Changing consistently with a transformation of the input; for example, predicted coordinates rotating with the input structure. Lesson 0.3
- ergodic hypothesis
- The assumption that a long enough simulation visits each region in proportion to its equilibrium probability, so time averages equal ensemble averages. Lesson 4.2
- eukaryote
- An organism whose cells keep their DNA in a nucleus and have other membrane-enclosed compartments: animals, plants, and fungi. Lesson 0.1
- evolution
- Change in the genetic makeup of populations over time, driven by mutation, natural selection, and genetic drift. Lesson 5.1
- exon
- Part of a gene kept in the final mRNA. Lesson 0.2
- expression tag
- Extra residues added to make a protein easy to purify, often left in the deposited sequence. Lesson 6.3
- extracellular
- Outside the cell, where secreted proteins often carry disulfide bonds. Lesson 0.1
- FASTA
- The standard text format for sequences: a header line starting with ">", then the residues on one or more lines. Lesson 2.1
- fibrous protein
- A protein built as long fibers from repeated sequence, such as collagen or keratin's coiled coils. Lesson 3.5
- fold class
- The coarsest grouping of folds by secondary structure, as all-α, all-β, α/β (helices and parallel strands interleaved), or α+β (helices and mostly antiparallel strands kept apart). Lesson 3.4
- folding funnel
- Folding pictured as descent on a funnel-shaped energy landscape, where most downhill moves lead toward the native state. Bumps on the walls are kinetic traps. Lesson 4.4
- force field
- A set of formulas and fitted parameters that computes a molecule's energy from its atom positions, used in simulation, minimization, and docking. Lesson 1.2
- formal charge
- Charge assigned by giving an atom all its unshared electrons and half its bonding electrons, then comparing with the neutral atom's valence count. It differs from a force-field partial charge. Lesson 0.4
- fraction bound
- The fraction θ of protein molecules with ligand bound, θ = [L]/(Kd + [L]) for a single site. Lesson 4.3
- frame-shift mutation
- An insertion or deletion not a multiple of three bases, which shifts the reading frame and scrambles every later codon. Lesson 5.1
- free energy
- G = H − TS. At fixed temperature and pressure, a negative change in G favors a process; the rate also depends on barriers. Lesson 1.5
- functional group
- A small group of atoms with characteristic chemistry, such as hydroxyl or carboxyl, which it lends to any molecule carrying it. Lesson 0.4
- gap penalty
- The cost of a gap. Affine penalties charge more to open a gap than to extend it, since one long indel is likelier than several short ones. Lesson 5.2
- gas constant
- R = 1.987 × 10⁻³ kcal/(mol·K), Boltzmann's constant per mole (R = N_A k_B). Lesson 4.1
- gene
- A stretch of DNA that is transcribed into RNA, usually one that encodes a protein. Lesson 0.2
- gene duplication
- An accidental copy of a gene within a genome. The copies then diverge, often in function; the globins arose this way. Lesson 5.1
- genetic code
- The table mapping the 64 codons to the 20 amino acids and stop; nearly identical in all organisms. Lesson 0.2
- genetic drift
- Random change in an allele's frequency between generations, regardless of its effect. It spreads or removes neutral variants. Lesson 5.1
- genome
- An organism's complete DNA. Lesson 5.1
- globular protein
- A water-soluble protein folded compactly, with a hydrophobic core and a polar surface. Lesson 3.5
- glycosidic bond
- The bond from a sugar's C1′ to its base: to N9 of a purine or N1 of a pyrimidine. Lesson 7.1
- glycosylation
- Attaching sugar chains to a protein, common on proteins that work outside cells. Lesson 3.5
- guanidinium
- Arginine's side-chain tip, one carbon bonded to three nitrogens that share a positive charge. Lesson 2.2
- hairpin loop
- Unpaired bases closed by one pair, capping the end of a stem. Lesson 7.2
- heat capacity
- Energy change per degree, C = Var(E)/kT². It peaks at a melting transition, sharply for a two-state protein. Lab 4
- heavy atom
- Any atom other than hydrogen. X-ray structures usually list only heavy atoms. Lab 0
- helical wheel
- A view down a helix's axis that places each residue 100° from the previous one, showing which residues share a face. Lesson 2.3
- heme
- A flat ring holding an iron atom. In hemoglobin and myoglobin it binds the oxygen. Lesson 3.6
- Henderson–Hasselbalch equation
- pH = pKa + log10([deprotonated]/[protonated]). It makes the protonated fraction a logistic function of pH. Lesson 1.3
- HETATM
- The PDB record type for atoms outside standard residues, such as water, ligands, metal ions, and modified residues. Lab 0
- Hill coefficient
- The exponent n in θ = pⁿ/(P50ⁿ + pⁿ), measuring how steep a binding curve is: 1 for independent sites, above 1 for positive cooperativity (about 2.8 for hemoglobin). Lesson 4.3
- homolog
- A protein or gene that shares a common ancestor with another. Homology is yes-or-no; similarity is measured. Lesson 5.1
- homology model
- A protein model built from a related protein's experimental structure, with no measurement of the protein itself. Lesson 6.1
- homology modeling
- Predicting a structure by adapting the known structure of a homolog (the template). Lesson 5.1
- homomer
- A complex of identical chains. A heteromer has different chains, like hemoglobin's α₂β₂. Lesson 3.6
- Hoogsteen edge
- One of a purine's three pairing edges, with the Watson–Crick and sugar edges. Lesson 7.2
- hormone
- A signal molecule, some of them proteins like insulin, carried through the body to cells with the matching receptor. Lesson 0.1
- HP model
- A lattice folding model with hydrophobic (H) and polar (P) residues, scoring −1 per H–H contact. Small enough to enumerate exactly. Lab 4
- hydrogen bond
- A directional interaction between a donor-bound hydrogen and a suitable acceptor. In proteins, donor and acceptor are commonly N or O, with favorable heavy-atom separations often around 2.7–3.2 Å. Lesson 1.4
- hydrolysis
- Splitting a bond by adding water across it, the reverse of condensation. It breaks proteins into amino acids and ATP into ADP. Lesson 4.1
- hydropathy
- A number per amino acid for how much its side chain prefers oil to water; positive is hydrophobic. Kyte and Doolittle's is the usual scale. Lesson 2.3
- hydrophilic
- Water-loving: polar or charged, forming hydrogen bonds or ionic interactions with water. Lesson 1.5
- hydrophobic
- Water-avoiding: nonpolar, mostly carbon and hydrogen, with nothing for water to hydrogen bond to. Lesson 1.5
- hydrophobic core
- A folded protein's interior, packed mostly with hydrophobic side chains away from water. Lesson 1.5
- hydrophobic effect
- The solvent-dependent tendency of nonpolar groups to reduce their exposure to water. Its free-energy balance includes the surrounding water and varies with solute size, temperature, and geometry. Lesson 1.5
- hydrophobic moment
- The vector sum of hydropathies, each pointing toward its residue on a helical wheel. Large when one face is oily and the other polar. Lesson 2.3
- hydrophobic residue
- A residue whose side chain is mostly carbon and hydrogen and can't hydrogen-bond to water (A, V, L, I, M, F, W). These tend to be buried. Lesson 2.2
- hydroxyl
- –OH: polar, donates and accepts hydrogen bonds. Found on serine, threonine, and tyrosine. Lesson 0.4
- hydroxyproline
- Proline hydroxylated after the chain is made. It stabilizes collagen's triple helix. Lesson 3.5
- IC50
- The inhibitor concentration that halves an activity. Related to Kd but dependent on the assay. Lesson 4.3
- imidazole
- Histidine's five-membered ring with two nitrogens. Its pKa near 6 lets it switch between neutral and positive. Lesson 2.2
- indel
- An insertion or deletion; a gap in an alignment. Lesson 5.1
- information content
- log₂ 20 minus a column's Shannon entropy: high where the family allows few residues. A sequence logo draws it as stack height. Lesson 5.3
- inhibitor
- A molecule that blocks an enzyme, often a drug occupying its active site. Lesson 3.4
- insertion code
- A letter after a residue number (60A, 60B) for residues inserted without renumbering the rest. Lesson 6.3
- intercalation
- A flat molecule slipping between two base pairs, lengthening and unwinding the helix. Lesson 7.1
- interface
- The surface where two chains or molecules touch, measured by the accessible area each loses. Lesson 3.6
- internal coordinates
- Bond lengths, bond angles, and torsions in place of x, y, z per atom. Lesson 3.1
- internal energy
- The energy E stored in a molecule's bonds and contacts. A force field computes it from coordinates and leaves out entropy. Lesson 4.1
- internal loop
- Unpaired bases on both sides between two stems. Lesson 7.2
- intrinsically disordered
- Having no single stable structure on its own. Many such proteins and regions fold only when they bind a partner. Lesson 3.5
- intron
- Part of a gene cut out of the RNA before translation. Lesson 0.2
- invariant
- Unchanged by a specified transformation; for example, pairwise distances under a rigid rotation. Lesson 0.3
- ion
- An atom or group with a whole-number net charge from gained or lost electrons (or protons). Lesson 1.1
- isoelectric point
- The pH at which a molecule's net charge is zero. Lesson 1.3
- isomer
- Molecules with the same atoms arranged differently: structural isomers differ in bond graph, cis–trans and optical isomers in 3D arrangement. Lesson 1.1
- isotope
- A form of an element with a particular number of neutrons; chemically almost identical to the others. Lesson 1.1
- kcal/mol
- Kilocalories per mole, the usual unit for molecular energies. 1 kcal = 4.184 kJ; a mole is 6.02 × 10²³ molecules. Lesson 1.1
- kilodalton
- 1,000 daltons. A residue averages about 110 Da, so 30 kDa is about 270 residues. Lesson 0.1
- kinase
- An enzyme that puts a phosphate on a protein, usually on Ser, Thr, or Tyr, switching the protein's behavior. Lesson 2.2
- kinetic trap
- A low-energy non-native state, separated from the native one by a barrier, where a folding chain or a simulation gets stuck. Lesson 4.4
- knowledge-based potential
- An energy term from how often contacts occur in known structures, by inverting the Boltzmann relation: E ≈ −RT ln(frequency). Lesson 4.3
- kT
- The thermal energy scale kBT for one molecule, or RT per mole: about 0.6 kcal/mol at room temperature. Comparing an energy or barrier with this scale helps assess populations or rates. Lesson 1.1
- L-amino acid
- L is the standard handedness of chiral amino acids incorporated by ribosomes; D is its mirror form. D residues occur in specialized pathways and some modified peptides; glycine is achiral. Lesson 1.6
- Lennard-Jones potential
- The 12-6 formula force fields use for van der Waals energy, a steep repulsive wall plus a shallow attractive well. Lesson 1.4
- Levinthal's paradox
- A random search of a chain's conformations would take longer than the age of the universe, yet proteins fold in milliseconds to seconds, so folding isn't a random search. Lesson 4.4
- ligand
- Any small molecule or ion that binds to a protein. Lesson 1.2
- lipid bilayer
- Two layers of lipids with their nonpolar tails facing inward and polar heads facing water; a common membrane arrangement about 4 nm thick. Lesson 0.1
- local resolution
- The resolution of one region of a cryo-EM map, which can be much worse than the map's overall figure. Lesson 6.1
- log-odds
- The log of a probability ratio, here how often a pair appears in true alignments over how often by chance. Summed, they score a likelihood ratio. Lesson 5.2
- lone pair
- A pair of valence electrons not assigned to a bond. Ordinary neutral alcohol oxygen has two lone pairs; an amine nitrogen usually has one. Lesson 0.4
- loop
- A stretch of chain with no regular structure: in a protein, backbone that is neither helix nor strand, often ordered and functional; in RNA, unpaired bases between or at the end of stems. Lesson 3.3
- low-complexity region
- A stretch built from few residue types, such as a run of glutamines; often disordered. Lesson 2.3
- macrostate
- A set of microstates treated as one state, such as "folded" or "bound". Its free energy sums over its members. Lesson 4.2
- major groove
- B-DNA's wider groove, where each base pair shows a distinct pattern that proteins read. Lesson 7.1
- marginal stability
- The small net free energy of folding, typically 5–15 kcal/mol, a difference of much larger enthalpy and entropy terms. One mutation can erase it. Lesson 4.4
- melting temperature
- The temperature at which half a protein's molecules are unfolded, where ΔG of folding is zero. Lesson 4.1
- membrane protein
- A protein embedded in or associated with a membrane. Regions facing lipid tails are usually nonpolar; a water-filled pore can have a polar lining. Lesson 0.1
- metal coordination
- Bonds between a metal ion and atoms that supply both shared electrons, such as the histidine nitrogen holding heme iron. Lesson 6.4
- methyl
- –CH₃, the smallest carbon group; nonpolar. Alanine's whole side chain. Lesson 0.4
- Metropolis Monte Carlo
- A sampler for the Boltzmann distribution: propose a random move, always accept it if the energy drops, and accept an uphill move with probability e^(−ΔE/kT). Lesson 4.2
- micrometer
- 10⁻⁶ m. Cells are 1–100 µm across. Lesson 0.1
- microstate
- One complete arrangement of a system with every atom placed; for a protein, one conformation. Lesson 4.2
- minor groove
- B-DNA's narrower groove, where base pairs look nearly alike; small drugs often bind there. Lesson 7.1
- missense mutation
- A DNA change that replaces one amino acid with another, such as the sickle-cell mutation in β-globin. Lesson 5.1
- mmCIF
- The PDB's master format: named tables (categories) such as _atom_site, with no size limits. Lesson 6.3
- modified nucleotide
- A nucleotide chemically altered after the RNA is made, such as pseudouridine (Ψ). Structure files give each its own residue name. Lab 7
- MODRES
- A structure-file record naming a modified residue and the standard amino acid it came from. Lesson 3.5
- molar
- Moles per liter, written M; 1 M is 6.02 × 10²³ molecules per liter. Lesson 1.3
- mole
- 6.022 × 10²³ of something, usually molecules. Quoting energies per mole scales them to laboratory amounts. Lesson 4.1
- molecular dynamics
- Moving atoms by integrating Newton's equations with force-field forces at a set temperature. It samples the Boltzmann distribution over time. Lesson 4.2
- molecular formula
- A count of a molecule's atoms, such as C₂H₆O, saying nothing about how they connect. Lesson 0.4
- molecular replacement
- Getting starting phases by placing a known similar structure (or a prediction) in the new crystal by rotation and translation searches. Lesson 6.1
- MolProbity
- The standard program for checking a structure's geometry, including Ramachandran outliers. Lesson 3.2
- monomer
- A small molecular building block that can be joined into a polymer, such as a nucleotide or amino acid. Lesson 0.2
- motif
- A small recurring pattern. In structure, a few secondary-structure elements such as a β-hairpin or helix-turn-helix; in sequence, a short conserved run such as a kinase's DFG. Lesson 3.4
- mRNA
- Messenger RNA: the copy of a protein-coding gene that the ribosome reads. Lesson 0.2
- multiloop
- A loop from which three or more stems leave. Lesson 7.2
- multiple sequence alignment
- Many related sequences stacked so that each column holds residues descended from the same ancestral position. Lesson 5.3
- mutation
- A change in DNA sequence: a substitution, insertion, or deletion. Most are neutral or harmful; a few help. Lesson 5.1
- mutual information
- How much one alignment column tells you about another; zero for independent columns. Shared ancestry and indirect chains inflate it. Lesson 5.4
- myoglobin
- Muscle's single-chain oxygen-binding protein, with the same globin fold and heme pocket as one hemoglobin chain. Lesson 3.6
- N-terminus
- The start of a protein chain, with the free amino group. Sequences are written and numbered from here. Lesson 2.1
- nanometer
- 10⁻⁹ m. A typical protein is a few nanometers across. Lesson 0.1
- native state
- A protein's folded, functional structure under normal conditions. Lesson 4.4
- natural selection
- Variants that help survival and reproduction becoming more common, and harmful ones disappearing. Lesson 5.1
- nearest-neighbor model
- A duplex's stability as a sum over neighboring base-pair steps, each with a measured free energy. Lesson 7.1
- Needleman–Wunsch
- Dynamic programming for the best global alignment, using every residue of both sequences. Lesson 5.2
- NeRF
- Natural extension reference frame: placing each next atom of a chain from the three before it, given a bond length, an angle, and a torsion. Lesson 3.1
- neutral theory
- Kimura's proposal that most molecular variation is selectively neutral and spreads by drift. Lesson 5.1
- NMR
- Nuclear magnetic resonance: nuclei in a strong magnetic field absorb radio waves at frequencies that reveal their chemical surroundings and nearby atoms. Lesson 6.1
- NOE
- Nuclear Overhauser effect: magnetization transfer between nuclei close in space, falling off as 1/r⁶, so an NOE puts two hydrogens within about 5–6 Å. Lesson 6.1
- noncanonical pair
- Any base pair other than Watson–Crick and G·U, classed by which edges meet and whether the glycosidic bonds are cis or trans. Lesson 7.2
- noncovalent interaction
- An interaction that leaves ordinary covalent connectivity unchanged, such as a hydrogen bond, ion pair, or van der Waals contact. Its energy depends on geometry and environment. Lesson 1.4
- nonpolar
- Having little charge separation, as in many hydrocarbon groups. Nonpolar groups still interact with water, including through dispersion forces. Lesson 1.2
- nucleic acid
- DNA or RNA, a polymer of nucleotides. Lesson 0.2
- nucleoside
- A base joined to a sugar, without the phosphate. Lesson 7.1
- nucleotide
- The monomer of DNA and RNA, made of a base, a five-carbon sugar, and a phosphate group. Lesson 0.2
- Nussinov algorithm
- Dynamic programming that finds the nested structure with the most base pairs in O(n³) time. Lesson 7.2
- occupancy
- The fraction of molecules in the crystal with an atom at the given position; alternate conformations share it. Lesson 6.2
- octet rule
- Atoms such as C, N, and O bond until they have eight outer electrons; hydrogen needs two. Lesson 1.1
- orbital
- The region around a nucleus where a given electron is likely to be found. Each holds at most two electrons. Lesson 1.1
- organelle
- A membrane-enclosed compartment inside a eukaryotic cell, such as the nucleus or a mitochondrion. Lesson 0.1
- ortholog
- Homologs split by a speciation: the same gene in two species, like human and whale myoglobin. Lesson 5.1
- oxidation
- Loss of electrons. Disulfide formation oxidizes two thiols; reduction reverses the process. The cytosol usually favors reduced thiols, while some internal compartments favor disulfides. Lesson 1.6
- P50
- The oxygen pressure at which a carrier is half saturated, the gas-phase equivalent of Kd: about 26 torr for hemoglobin and 2.8 torr for myoglobin. Lesson 4.3
- PAE
- AlphaFold's expected error in one residue's position when the model is aligned on another. It shows how confidently domains are placed relative to each other. Lesson 6.1
- pairwise alignment
- A residue-by-residue correspondence between two sequences, with gaps, chosen to maximize a score. Found by dynamic programming. Lesson 5.2
- parallel sheet
- A sheet whose neighboring strands run in the same N→C direction. Lesson 3.3
- paralog
- Homologs split by a gene duplication, like myoglobin and the hemoglobin chains. Lesson 5.1
- partial charge
- A fraction of an electron's charge on an atom in a polar bond (δ⁺ or δ⁻). Force fields assign one to every atom. Lesson 1.2
- partial pressure
- One gas's share of a mixture's pressure, used as its concentration; in mm Hg or torr, which are the same unit. Lesson 4.3
- particle (cryo-EM)
- One copy of the molecule caught in the ice, in some orientation, imaged as a noisy 2D projection. Lesson 6.1
- partition function
- The normalizer of the Boltzmann distribution, Z = Σ e^(−E/kT). Its log gives the free energy, F = −kT ln Z. Lesson 4.2
- PDB
- The Protein Data Bank, an archive of experimentally determined molecular structures and associated information. An identifier such as 1UBQ selects a deposited entry. Lesson 0.3
- PDB format
- The original fixed-column text format for structures, with fields defined by column ranges, not spaces. Lesson 6.3
- pentose
- A five-carbon sugar; ribose in RNA and deoxyribose in DNA. Lesson 7.1
- peptide bond
- The C–N bond from one residue's carboxyl carbon to the next residue's amino nitrogen. Partial double-bond character keeps it flat and unable to rotate. Lesson 2.1
- Pfam
- A database of over 20,000 protein families, each with a curated seed alignment, a profile HMM, and a full alignment of members from UniProt. Lesson 5.3
- pH
- Minus the base-10 logarithm of proton activity, approximated by concentration relative to 1 M in dilute solution. Lowering pH by one unit increases proton activity tenfold. Lesson 1.3
- phase problem
- A detector records each reflection's intensity but not its phase, and the phases decide what the density map looks like. Lesson 6.1
- phenyl
- A benzene ring attached by one carbon, –C₆H₅. Phenylalanine's side chain is CH₂ plus phenyl. Lesson 0.4
- phosphate
- A group containing phosphorus bonded to four oxygens. Its attachments and protonation set its charge; phosphate groups link nucleic-acid backbones and can modify protein behavior. Lesson 0.4
- phosphodiester bond
- The link between nucleotides: a phosphate joining one sugar's 3′ carbon to the next sugar's 5′ carbon. Lesson 7.1
- phospholipid
- A lipid with a charged, water-loving head and two oily fatty-acid tails; the main component of membranes. Lesson 0.1
- phosphorylation
- A kinase adding a phosphate to Ser, Thr, or Tyr: about two negative charges, often switching the protein on or off. Lesson 3.5
- phylogenetic tree
- A tree of descent: sequences at the leaves, ancestors at internal nodes, branch lengths showing amount of change. Lesson 5.1
- physiological pH
- About 7.2–7.4, the pH inside cells and in blood. "Charged" without qualification means charged at this pH. Lesson 1.3
- pKa
- The pH at which a group is protonated half the time; below it mostly protonated, above it mostly not. Lesson 1.3
- pLDDT
- AlphaFold's per-residue confidence, 0 to 100, stored in the B-factor column. High is good, the opposite sense of a B factor. Lesson 6.1
- polar
- Sharing electrons unevenly, leaving partial positive and negative ends. Polar groups attract water and each other. Lesson 1.2
- polar residue
- A residue with a neutral side chain that can donate or accept hydrogen bonds (S, T, N, Q, and by some counts Y and C). Lesson 2.2
- polymer
- A molecule built by linking many smaller units, called monomers. Lesson 0.2
- polypeptide
- A chain of amino acids joined by peptide bonds. A protein is one or more polypeptides, folded. Lesson 2.1
- polyproline II
- An extended, left-handed helical conformation near (−75°, +145°), common in loops and unfolded chains. Lesson 3.2
- position-specific scoring matrix
- A profile turned into scores: residue a at position i scores log(q_i(a)/p_a), a different log-odds column at every position. Lesson 5.3
- post-translational modification
- A chemical change to a protein after it is made, such as an added phosphate or sugar. The encoded sequence is unchanged. Lesson 3.5
- Potts model
- A Boltzmann distribution over sequences with a field for each position and a coupling for each pair, fit to reproduce an alignment's single and pair frequencies. Lesson 5.4
- primary structure
- A protein's amino-acid sequence, read from N- to C-terminus. Lesson 2.1
- profile hidden Markov model
- A profile with match, insert, and delete states and transition probabilities between them. Aligning to it uses dynamic programming (Viterbi). Lesson 5.3
- prokaryote
- A bacterium or archaeon, whose cells lack a membrane-enclosed nucleus. Lesson 0.1
- protein fold
- A chain's overall 3D architecture: the arrangement and connection order of its helices and strands. Lesson 3.4
- protein language model
- A neural network trained to fill in masked residues across hundreds of millions of protein sequences. It learns family constraints, including coevolution, implicitly. Lesson 5.1
- protonation state
- Which protons are attached to an ionizable group. Gaining one proton raises its net charge by one; the resulting charge also depends on the starting group. Lesson 1.3
- pseudocount
- A few pretend observations added to alignment counts so unseen residues get small probabilities, not zero. Lesson 5.3
- pseudoknot
- Pairs that cross other pairs, so two stems interleave. They break the nesting that folding algorithms assume. Lesson 7.2
- pseudorotation phase
- The angle P summarizing a five-membered ring's pucker from its five ring torsions. North (near 18°) is C3′-endo, south (near 160°) C2′-endo. Lesson 7.1
- purifying selection
- Selection removing harmful changes, which keeps important positions, such as buried and active-site residues, conserved. Lesson 5.1
- purine
- A base with two fused rings, adenine (A) or guanine (G). Lesson 0.2
- pyrimidine
- A base with a single ring, cytosine (C), thymine (T), or uracil (U). Lesson 0.2
- Q3
- The fraction of residues whose three-state secondary-structure label (helix, strand, coil) a prediction gets right. Lesson 3.3
- quaternary structure
- The arrangement of several chains in a complex. Lesson 3.4
- R group
- "The rest of the molecule" in a drawing; in an amino acid, the side chain. Lesson 0.4
- R-factor
- The relative disagreement between measured and model-computed structure-factor amplitudes. About 0.20 is typical for proteins. Lesson 6.2
- R-free
- The R-factor on 5–10% of reflections left out of refinement: a validation error that exposes overfitting. Lesson 6.2
- radius of gyration
- The root-mean-square distance of atoms from their center: one number for size and compactness. Lab 0
- Ramachandran plot
- ψ plotted against φ for a protein's residues, showing which backbone conformations are allowed and which push atoms into each other. Lesson 3.2
- reading frame
- One of three ways to split a sequence into codons, set by the starting base. Lesson 0.2
- receptor
- A protein that binds a signal molecule, such as a hormone, and changes shape in response. Lesson 0.1
- reducing agent
- A chemical that donates electrons, here turning disulfide bonds back into free –SH groups. β-Mercaptoethanol is the classic one. Lesson 4.4
- refinement
- Adjusting a model's coordinates, B factors, and occupancies to fit the diffraction data, with ideal bond geometry as a restraint. Lesson 6.1
- reflection
- One spot of a diffraction pattern, corresponding to one Fourier coefficient of the electron density. Lesson 6.1
- relative solvent accessibility
- A residue's accessible area over its maximum possible. Below about 0.2 the residue is usually called buried. Lesson 2.4
- residue
- One monomer in a chain: an amino acid in a protein, or a nucleotide in DNA or RNA. Joining the chain costs each a water molecule, hence the name. Lesson 2.1
- resolution
- The finest spacing of detail a structure determination resolved, in ångströms. Smaller is better; side chains are clear near 2 Å. Lesson 6.2
- resonance
- Electrons that no single drawing of single and double bonds captures, so the real bonds are in between, as in a carboxylate or the peptide group. Lesson 1.1
- restraint
- A bound, such as a maximum distance between two atoms, that a model must satisfy during calculation or refinement. Lesson 6.1
- reverse complement
- A sequence's partner strand read 5′ to 3′: complement each base and reverse the order. Lesson 7.1
- ribose
- RNA's pentose, with a hydroxyl (2′-OH) on its 2′ carbon. Lesson 7.1
- ribosome
- A complex of RNA and proteins that reads mRNA and joins amino acids into a protein chain. Lesson 0.2
- riboswitch
- A stretch at the start of an mRNA that binds a small molecule and changes shape, turning its gene's protein production up or down. Lesson 7.2
- ribozyme
- An RNA that catalyzes a chemical reaction. Lesson 7.2
- rigid motion
- Rotating and translating a whole molecule: every coordinate changes, no distance does, so the molecule is unchanged. Lesson 0.3
- RMSD
- Root-mean-square deviation between matching atoms of two structures after the best rigid superposition. Under 1 Å is essentially the same structure. Lesson 3.1
- RNA
- Ribonucleic acid: a usually single-stranded nucleotide polymer over A, C, G, and U. mRNA, tRNA, and rRNA carry out protein synthesis. Lesson 0.2
- rRNA
- The ribosome's RNA, which forms most of its structure and its catalytic site. Lesson 7.2
- salt bridge
- An interaction between oppositely charged groups, often identified by a positive N and negative O within about 4 Å. Its contribution to stability depends on solvation and the surrounding structure. Lesson 1.4
- sampling problem
- A simulation failing to visit all important states in the time available, usually trapped by barriers. Often a bigger limit than the energy function. Lesson 4.2
- SCOP
- A database that classifies protein domains by hand into class, fold, superfamily, and family. Lesson 3.4
- scoring function
- A fast estimate of binding free energy used to rank docked poses and compounds. Typical errors are about 2 kcal/mol, a factor of 30 in Kd. Lesson 4.3
- second law of thermodynamics
- Total entropy never decreases. Local order, such as a folded protein, is paid for by heat released to the surroundings; ΔG < 0 is the law stated for the system alone. Lesson 4.1
- secondary structure
- The regular local shapes of a protein backbone, chiefly α-helices and β-strands, held by backbone hydrogen bonds. Lesson 3.3
- secondary structure (RNA)
- The set of base pairs an RNA forms with itself, drawn as stems and loops. Lesson 7.2
- SEQRES
- Records of the full sequence of each polymer studied, including residues with no coordinates. Lesson 6.3
- sequence identity
- The fraction of aligned positions where two sequences have the same residue. Lesson 5.1
- sequence logo
- A profile drawn as letter stacks: each column's height is its information content, and each letter's height its frequency. Lesson 5.3
- sequence profile
- Per-column amino-acid probabilities computed from a multiple sequence alignment. Lesson 5.3
- sequence weighting
- Down-weighting near-duplicate sequences (often by 1 over the number of neighbors above 80% identity) so clusters don't dominate. The weights sum to the effective number of sequences. Lesson 5.3
- side chain
- The part of an amino acid that varies, attached to Cα. It sets the residue's size, charge, and preference for water or oil. Lesson 2.1
- signal peptide
- A hydrophobic stretch at a new protein's N-terminus that steers it to a membrane and is then cut off. Lesson 2.3
- simulated annealing
- Monte Carlo cooled slowly from a high temperature, crossing barriers early and settling into a low minimum late. An optimizer, not a sampler. Lesson 4.2
- skeletal formula
- A bond drawing with carbon symbols and their hydrogens mostly omitted. Unlabeled corners and endpoints are carbons; a labeled endpoint is the atom named there. Lesson 0.4
- Smith–Waterman
- Dynamic programming for the best local alignment, the highest-scoring pair of segments; a floor of zero cuts off poor stretches. Lesson 5.2
- solvation
- Interactions of a molecule or ion with its surrounding solvent. Desolvation removes or replaces these interactions, as when a group becomes buried. Lesson 1.4
- solvation term
- An energy-function term that stands in for unmodeled water, including the hydrophobic effect. Lesson 4.4
- solvent-accessible surface
- The surface traced by the center of a probe sphere (1.4 Å, the size of a water molecule) rolled over a molecule's atoms. Lesson 2.4
- solvent-excluded surface
- The boundary of the space a rolling probe can't enter, hugging the atoms and closing over crevices too narrow for water; what viewers draw as "the surface". Lesson 2.4
- space group
- The symmetry operations that relate copies of the molecule within a crystal. Lesson 6.2
- space-filling
- Each atom drawn as a sphere of its van der Waals radius, showing shape and hiding the interior. Lesson 6.4
- species
- Organisms that can interbreed. Once two populations stop interbreeding, their genomes diverge independently. Lesson 5.1
- spin
- The property that makes some nuclei, including hydrogen's, behave like tiny magnets. Lesson 6.1
- standard state
- The reference conditions (1 M, 1 atm) that make an equilibrium constant a pure number. The ° in ΔG° marks them. Lesson 4.3
- start codon
- A codon used to initiate translation and establish a reading frame, usually AUG. Cellular signals and surrounding sequence help determine which start is used. Lesson 0.2
- static disorder
- Molecules of a crystal frozen in slightly different positions. It raises B factors just as motion does. Lesson 6.2
- stem
- A run of consecutive base pairs in an RNA, forming a short A-form helix. Lesson 7.2
- steric clash
- Nonbonded atoms closer than their van der Waals radii allow: very costly, and a sign of model error. Lesson 1.4
- steric hindrance
- Atoms in each other's way. A conformation that pushes atoms closer than their sizes allow is forbidden. Lesson 3.2
- stop codon
- UAA, UAG, or UGA: ends translation and releases the chain. Lesson 0.2
- structural bioinformatics
- The part of bioinformatics that represents, stores, analyzes, and predicts the 3D structures of biomolecules. Lesson 0.3
- structural formula
- A drawing of a molecule showing every atom and every bond. Lesson 0.4
- structure determination
- Building an atomic model from experimental data such as X-ray diffraction, NMR, or cryo-EM. Lesson 0.3
- structure factor
- The complex Fourier coefficient F(h) behind one reflection: an amplitude, which is measured, and a phase, which is not. Lesson 6.1
- structure prediction
- Computing a molecule's 3D coordinates from its sequence, without an experiment. Lesson 0.3
- substitution matrix
- Scores for aligning each pair of amino acids: the log-odds of how often relatives pair them versus chance. Lesson 5.2
- substrate
- The molecule an enzyme acts on. Lesson 5.3
- subunit
- One part of a multi-part assembly: a chain in a multi-chain protein, or one of the ribosome's two RNA–protein halves. Lesson 3.6
- sugar pucker
- Which ring atom of the sugar sits out of plane, and on which side: C2′-endo in B-DNA, C3′-endo in RNA. Lesson 7.1
- syn
- The glycosidic orientation with the base turned over the sugar; seen in Z-DNA's guanines. Lesson 7.1
- synonymous mutation
- A codon change that leaves the encoded amino acid unchanged. It can still affect RNA processing or translation. Lesson 0.2
- synonymous substitution
- A DNA change that keeps the encoded amino acid, usually at a codon's third position. Selection barely sees it. Lesson 5.1
- T state
- Hemoglobin's two quaternary arrangements, T (tense, deoxy, low oxygen affinity) and R (relaxed, oxy, high affinity). Lesson 3.6
- tautomer
- One of interconvertible chemical forms differing in proton positions and associated bonding, such as the two neutral histidine ring forms. Lab 1
- template strand
- The DNA strand RNA polymerase reads; the RNA is complementary to it. Lesson 0.2
- tertiary contact
- A contact between distant parts of a folded RNA that holds its 3D shape, often noncanonical and crossing the secondary structure. Lesson 7.2
- tertiary structure
- The 3D fold of one whole chain. Lesson 3.4
- tetramer
- A complex of four chains. A dimer has two. Lesson 3.6
- thermodynamic hypothesis
- Anfinsen's conclusion from refolding ribonuclease: under physiological conditions the native structure is the lowest-free-energy state, and the sequence determines it. Lesson 4.4
- thiol
- –SH, the tip of cysteine's side chain. Two can join as a disulfide, –S–S–. Lesson 0.4
- topology
- The order, pairing, and direction of a protein's helices and strands. Lesson 3.4
- trans
- Opposite sides: the two Cα atoms across a peptide bond (ω near 180°, the usual form), matching groups across a double bond, or the two glycosidic bonds of a base pair. Lesson 2.1
- transcription
- Copying DNA into RNA, done by RNA polymerase. Lesson 0.2
- transcription factor
- A protein that switches a gene on or off by binding DNA next to it. Lesson 7.1
- translation
- Building a protein from an mRNA's codons, done by the ribosome with tRNAs as adaptors. Lesson 0.2
- transmembrane
- Crossing a membrane. A transmembrane helix is about 20 mostly hydrophobic residues, enough to span the bilayer's oily core. Lesson 2.3
- transporter
- A membrane protein that lets a particular molecule or ion cross the membrane. Lesson 0.1
- tRNA
- Transfer RNA: a small folded RNA that carries one kind of amino acid and pairs with the matching mRNA codon. Lesson 0.2
- twilight zone
- About 20–40% sequence identity, where homologs and chance matches overlap. Below about 20% (the midnight zone) homology is hard to detect from sequence. Lesson 5.1
- two-state folding
- Each molecule is either essentially folded or unfolded, with intermediates too rare to detect. Typical of small proteins. Lesson 4.4
- UniProt
- The main protein-sequence database. Its positions are a protein's reference numbering. Lesson 6.3
- unit cell
- The repeating box of a crystal; stacked copies of it make the whole crystal. Lesson 6.1
- urea
- A small polar molecule that at high concentration unfolds proteins by competing for their hydrogen bonds and solvating their interiors. Lesson 4.4
- valence
- An atom's combining capacity in a specified chemical state. Common neutral groups have H making one bond, O two, N three, and C four, with double bonds counting twice. Lesson 0.4
- valence shell
- The outermost electron shell, whose electrons form bonds. Lesson 1.1
- validation report
- The PDB's per-entry assessment of geometry, clashes, Ramachandran and side-chain outliers, and fit to the data, ranked against similar entries. Lesson 6.2
- van der Waals force
- A weak, short-range attraction between any two atoms from fluctuating electron clouds, a fraction of a kT per contact. Lesson 1.4
- van der Waals radius
- An atom's effective size in nonbonded contacts, about 1.7 Å for carbon. Atoms closer than the sum of their radii overlap. Lesson 1.4
- vitrified ice
- Water frozen so fast that it becomes a glass instead of ice crystals, keeping molecules in their solution shapes. Lesson 6.1
- Watson–Crick pair
- The canonical pairs A–T (A–U in RNA), with two hydrogen bonds, and G–C, with three. Lesson 7.1
- wedge bond
- A solid wedge is a bond toward the viewer, a dashed wedge one pointing away; plain lines lie in the page. Lesson 0.4
- wild type
- The common, normal version of a gene or protein, the reference for describing mutants. Lesson 5.1
- wobble pair
- The G·U pair, with two hydrogen bonds; common in RNA stems and at the third codon position. Lesson 7.2
- X-ray crystallography
- Determining a structure from the diffraction spots of X-rays passed through a crystal, by computing the electron density. Lesson 6.1
- Z-DNA
- A left-handed DNA helix formed by alternating C and G, with syn guanines and a zigzag backbone. Lesson 7.1
- zinc finger
- A small DNA-binding domain: a β-hairpin and a helix held together by a zinc ion bound to cysteines and histidines. Lesson 3.5
- Zuker algorithm
- The same recursion as Nussinov's, minimizing nearest-neighbor free energy instead of counting pairs. Lesson 7.2
- zwitterion
- A neutral molecule carrying both a positive and a negative charge, as a free amino acid does in water (–NH₃⁺ and –COO⁻). Lesson 2.1
- α-helix
- A right-handed spiral of the backbone with 3.6 residues and 5.4 Å per turn, held by a hydrogen bond from each C=O to the N–H four residues ahead. Lesson 3.3
- αL region
- The small region near (+57°, +47°), a left-handed helical conformation used mostly by glycine. Lesson 3.2
- αR region
- The Ramachandran region near (φ, ψ) = (−63°, −43°), used by right-handed helices. Lesson 3.2
- β-hairpin
- Two antiparallel strands joined by a short turn. Lesson 3.3
- β-sheet
- Two or more β-strands side by side, joined by backbone hydrogen bonds, with side chains alternating above and below. Lesson 3.3
- β-strand
- A nearly fully extended stretch of backbone. Strands pair side by side through hydrogen bonds into sheets. Lesson 3.3
- β-turn
- A four-residue reversal of the chain, closed by a hydrogen bond from the C=O of the first residue to the N–H of the fourth. Lesson 3.3
- ΔΔG
- The change in a free energy of folding or binding caused by a mutation. Sign conventions vary, so check which one is used. Lesson 4.4
- π-helix
- A wider helix, about 4.3 residues per turn with i → i+5 hydrogen bonds; usually a one-turn bulge in an α-helix. Lesson 3.3
- φ
- The backbone torsion about the N–Cα bond, defined by C(i−1), N, Cα, C. Lesson 3.1
- χ1
- The first side-chain torsion angle, about the Cα–Cβ bond. Lesson 6.4
- ψ
- The backbone torsion about the Cα–C bond, defined by N, Cα, C, N(i+1). Lesson 3.1
- ω
- The torsion about the peptide bond: nearly always near 180° (trans). About 5% of bonds before a proline are cis, near 0°. Lesson 2.1