Tool Set Documentation

Revision as of 21:11, 11 August 2006 by Feig (talk | contribs)
(diff) ←Older revision | view current revision (diff) | Newer revision→ (diff)
Jump to: navigation, search


usage: [options] refPDB cmpPDB
options: [-out ca|cab|c|o|n|side|back|all...]
         [-detailed] [-chains]
         [-l min:max[...]]
         [-fit] [-fitxl]
         [-fitl min:max[...]] [-fitx min:max[...]]
         [-fitsel cab|ca|cb|heavy]
         [-resnumonly] [-useseg]
         [-align fasta]

Show source


This script calculates root mean square deviations between two structures. By default only the total RMSD of all (heavy) atoms is printed out. Other output modes can be selected with -out. The following options are recognized: ca (RMSD of C-alpha atoms), cb (C-beta atoms), cab (C-alpha and C-beta atoms), c (peptide backbone carbonyl atoms), o (peptide backbone carbonyl oxygen atoms), n (peptide backbone nitrogen), side (side chain heavy atoms), back (back bone heavy atoms), and all for all (heavy) atoms. With all of these options only a single value is reported. Alternatively, a complete set of RMSD values for different atom types, backbone, sidechains, residue types as well as each residue can be requested with -detailed. If multiple chains are present, the option -chains can be used to request RMSD values for each chain. Otherwise averages for all chains will be written out.
In most cases it is probably desirable to perform a least squares fit between the two structures before calculating root mean square deviations. While this is not done by default such a fit can be requested with -fit. The option -fitsel controls which atoms are used for the fit. Choices are cab (C-alpha and C-beta, default), ca (C-alpha), cb (C-beta), and heavy (all heavy atoms). The least squares fit can be restricted to a residue subset given with -fitl or to all residues outside a subset given with -fitx.
For loop modeling the RMSD calculation can be restricted to a residue subset with -l. If this option is specified -fitxl is available to automatically fit all residues outside the loop or fragment.
The option -nowarn is available for suppressing warnings about missing atoms. As with <docmark></docmark> atoms in the two structures are normally matched by atom name, residue name, and residue number. This would not allow the comparison of structures with different amino acid sequences. This is possible when the option -resnumonly is given. In this case, atoms are matched only by atom name and residue number while the residue names are ignored. This is particulary useful for calculating root mean square deviations of backbone atoms between homologous structures. This option also automatically turns off all warnings about missing atoms.


usage information

Examples 1vii.exp.pdb 1vii.exp.min.pdb
calculates the root mean square displacement of all (heavy) atoms between two structures

   1.6750 all -out ca -fit 1vii.exp.pdb 1vii.exp.min.pdb
calculates the C-alpha root mean square diesplacement between two structures after fitting them with a least squares fit procedure

   1.2460 CA -l 10:21 -fitxl 1vii.exp.pdb 1vii.exp.min.pdb
calculates the root mean square displacement for residues 10 through 21 between two structures. The structures are fitted with a least squares procedure of all residues outside the range for which the RMSD is calculated. This is a typical example for comparing loop structures.

   1.5776 all -detailed -l 10:15 -fitl 10:21 -fitsel ca 1vii.exp.pdb 1vii.exp.min.pdb
calculates the root mean square displacement for residues 10 through 15 between two structures after least squares fitting C-alpha atoms of residues 10 through 21. The detailed output contains RMSD values for each residue, residue type, and specific atom subsets.

all         1.5541
 back       1.0124
  CA        0.9523
  N         0.6613
  C         0.8758
  O         1.4102
 side       1.9508
  CAB       1.2323
  CB        1.5009
  ARG       1.7635