PROCHECK Operating Manual
PROCHECK Operating Manual
The first of the programs is called CLEAN. Its purpose is to produce a cleaned-up version of your coordinates file. The new file is given a .new extension and is used by the other programs in the suite for the calculation of the stereochemical parameters.
The cleaning-up process involves a number of checks. The first one ensures that the atoms in your structure have been correctly labelled in accordance with the IUPAC naming conventions (IUPAC-IUB Commission on Biochemical Nomenclature, 1970).
A typical error is that the Neta1 and Neta2 atoms of arginine are labelled the wrong way round. This error is corrected by the program. Also corrected are atom labels for Phe, Tyr, Asp and Glu residues. Thus, for example, the carbon atom in tyrosine labelled Cdelta1 is the one that gives the lowest chi-2 torsion angle.
The program switches atom labels when they are in error and shows the old atom names in a column at the right of the .new file.
The program also checks that the correct L/D stereochemical assignments have been made on individual residues, and corrects any errors found.
All other errors are merely flagged in the .new file and described in the clean.log file. Also listed in clean.log are: chain breaks, unknown residues, and missing atoms. Chain breaks are defined as locations where the distance between two adjacent Calpha atoms is greater than 5.0Å. Residue atoms on one side of such a break are marked with a `>' in the .new file while the atoms of the residue on the other side are marked with a `<'.
Atoms with alternate locations have only the location with the highest occupancy retained in the .new file. The alternate positions are written to the .new file, but are flagged: ATOM records are marked as ATALT and HETATM records as HEALT. Atoms with zero occupancy are flagged as ATZERO and HEZERO. None of the flagged atoms are included in the stereochemical checks.
Finally, residues with unusual values of the "notional" zeta dihedral angle (defined by the four atoms: Calpha-N-C-Cbeta) are marked with an asterisk in the .new file. The value of this angle should be > 23 degrees and < 45 degrees.
The second program in the suite is SECSTR which calculates all the required torsion angles, main-chain hydrogen-bond energies, and secondary structure assignments. This information is written out to the .rin file for use by the plotting programs.
The hydrogen-bond energies and the secondary structure assignments are calculated using the method of Kabsch & Sander (1983).
The third program in the suite is called NB. It identifies all non-bonded interactions between different pairs of residues in the protein structure. These are defined where the closest atom-atom contact between two residues is less than 2.0Å and the atoms concerned are 4 or more bonds apart.
All such closest contacts are written out to the .nb file for use by the plotting programs.
Note that, only intra-chain contacts are stored - inter-chain contacts are not considered.
The fourth program in the suite is called ANGLEN. It calculates all the main-chain bond lengths and bond angles in the structure. These are written out to the .lan file. Also calculated are the RMS distances from a best-fit plane for all planar sidechain groups. These comprise the aromatic rings of Phe, Tyr, Trp and His, and the end-groups of Arg, Asn, Asp, Gln and Glu. The coordinates of the atoms in the planar groups are written out to the .pln file.
The last 3 programs in the suite are the plotting programs which create the plots and the residue-by-residue listing for your structure. The plots and listing have a number of user-definable options which are held in the parameter file procheck.prm and which can be changed simply by editing the file using any text editor (see Customizing the PROCHECK plots). The plots are described in Sample plots, and the print-out is described in Appendix D.
PROCHECK Operating Manual