AtomSets

Predefined Atom Sets

Jmol recognizes and uses several keywords or tokens for several purposes: commands in the scripting language, colors, etc. Among them, there are keywords for predefined atom sets:

By element names:

carbon, oxygen, hydrogen, sulphur, etc.

Parts in proteins:

Backbone

Inclusion in this set is determined by atom id*, as follows:

• Peptide bond: N, H (bound to N), CA (alpha carbon), HA (bound to CA), C (carbonyl carbon), O or O1 (bound to C)
• In glycine, the two equivalent H are both in the backbone set: either H1 and 1HA or 1HA and 2HA.

• Termini:
  • OXT : second carbonyl oxygen on C-terminus
  • 1H, 2H, 3H : terminal amino hydrogens

(*) on PDB format: atom id is called atom name, and must be in these positions/columns:

• 13-14 : Chemical symbol, right justified, except for hydrogen atoms
• 15 : Remoteness indicator (alphabetic); e.g., in amino acid residues, alpha = A, beta = B, gamma = G, delta = D, epsilon = E.
• 16 : Branch designator (numeric).

Sidechain

Defined as (not backbone).

Parts in nucleic acids:

Backbone

Inclusion in this set is determined by atom id*, as follows:

• Phosphate groups:
  • phosphorus: P
  • oxygens bound to phosphorus: O1P, O2P

• Atoms in pentose:
  • carbon ring: C1', C2', C3', C4', C5'
  • hydrogens attached to carbon ring: H1', 1H2', 2H2' (only DNA), H3', H4', 1H5' and 2H5'
  • hydroxyls: O2', O3', O4', O5', 2HO' (H on 2'-hydroxyl, only RNA) (the ring oxygen is denoted O4, not O1).

Note: PDB files label pentose atoms with asterisk instead of prime signs. How does Jmol cope with this? Not much of a trouble: given the asterisk is a wildcard, "select C3*" will get pentose carbons either labeled with prime or asterisk!.

• Termini:
  • O5T : 5'-terminus oxygen (no phosphate)
  • H5T : 5'-terminus hydrogen (attached to O5T or O5')
  • H3T : 3'-terminus hydrogen (on 3'-hydroxyl)

• Atoms in bases:
  • ring, both purines and pyrimidines: N1, C2, N3, C4, C5, C6
  • ring, purines: N7, C8, N9
  • ring, pyrimidines: O2
  • substituents on ring:
    • N4 unique to C
    • N2 unique to G
    • N6 unique to A
    • C5M unique to T
    • O6 in G and I
    • O4 in T and U
    • S4 unique to thio-U

(*)Note: on PDB format, atom id is called atom name, and must be in these positions/columns:

• 13-14 : Chemical symbol, right justified, except for hydrogen atoms
• 15 : Remoteness indicator (alphabetic).
• 16 : Branch designator (numeric).

Sidechain

Defined as (not backbone).

Bases

Synonim of sidechain.

By type of molecule:

protein, nucleic, dna, rna, water, solvent, ligand...

By type of residue:

Inclusion in this sets is determined by residue id (only as far as it is written in the adequate field in the molecular coordinate file, usually PDB format).

Residue IDs:

• Nucleotides: A, G, C, T, U
• Amino acids: the 3-letter standard abbreviation

Residue sets:

• Nucleotides: purine, pyrimidine
• Amino acids: polar, nonpolar, charged, acidic, negative, basic, positive, aromatic, ...

Contributors

AngelHerraez, NicolasVervelle, Pimpim, JimHu, Miguel