NAME Psa - protein surface accessibility SYNOPSIS psa file DESCRIPTION Psa is a program for calculating the surface accessibility of all the atoms and residues from a PDB format coordinate file. It is recommended that you preprocess your coordinates with the program pdb2atm(1) to ensure they have a standard format. The program implements the method of Lee and Richards. By default HETATM groups, excluding water, are included in the calculations (see below as to how to control this). There are a number of data files that are read at run-time that control what the program does. The first of these is psa.dat. This can either be in the current directory or in the psa library directory. The file in the current directory is used first. This file contains a list of residue/group names, atom names and atomic radii. Within this library the hash symbol signifies a comment. The format of this library is the name of the residue/hetatm group (A3), then the number of atoms within this group (I3), note that this number should include the terminal oxygen of the residue. There then follows a list of the atom names and the corresponding Van der Waal's radii (A4,F6.2). For every residue it is wise to add an OXT record, otherwise you will get spurious error messages. Thus an example entry for alanine is: ALA 6 # alanine N 1.65 CA 1.87 C 1.76 O 1.40 CB 1.87 OXT 1.40 A number of options can be used to control psa. These are: -t, Send output to stdout. -pN.n, where N.n is a real number representing the probe radius, which is to be `rolled over the surface', 1.4A is half the inter-oxygen distance in ice, and so is associated with the Van der Waal's radius of water (and all oxygen atoms in general). -eN.n, where N.n is a real number representing the integration step, this is the plane spacing (in A) used to calculate the accessibility. The smaller this value the more accurate the results will be, but the slower the program will run. The default value seems to be a reasonable compromise. -a, this controls whether atomic accesibilities are written to file, if ON then a file with a .sol extension is produced, containing a PDB format coordinate set but with the B-value column containing the absolute atom accessibility for each atom. The default for this is false. -a, this controls whether residue accesibilities are written to file, if ON, then a file with a .psa extension is produced. Standard accesibilities were calculated in an A-X-A peptide (where X is the residue in question), in an extended conformation (o=-140, u=135 and w=180) with the exception of proline. Where applicable sidechain torsions were all trans. An important point to note is that in the summing of accessibilities into mainchain and sidechain values, the Ca atom is considered to be part of the sidechain, and is not considered in the mainchain values; this is so glycine can be treated consistently with the other amino acids. For further notes on this see the file standard.doc. The standard accessibility data was kindly provided by Simon Hubbard. Input can come from stdin if you want it to. An example of a .psa file is: # produced by psa, version 1.0a # # File of summed (Sum) and % (per.) accessibilities # probe radius : 1.400 # integration step : .050 # water included : F # hetatom included : T # accessibility type : CONTACT # # Res Res All atoms Non P side Polar Side Total Side Main Chain # Num type Sum Per. Sum Per. Sum Per. Sum Per. Sum Per. ACCESS 16 ILE .56 1.0 .01 .0 .00 .0 .01 .0 .55 5.6 ACCESS 17 VAL 4.17 8.7 4.17 11.0 .00 .0 4.17 11.0 .00 .0 ACCESS 18 GLY 12.72 53.6 8.20 75.7 .00 .0 8.20 75.7 4.52 35.0 ACCESS 19 GLY 9.61 40.4 .64 5.9 .00 .0 .64 5.9 8.96 69.4 ACCESS 20 TYR 30.63 49.3 28.82 68.1 1.81 17.1 30.63 57.9 .00 .0 ACCESS 21 THR 21.55 51.5 12.98 52.1 3.65 52.0 16.63 52.1 4.92 49.6 The units for the SUM columns are A2 and % for the Per columns. It is important to note that the Ca atom is considered part of the sidechain. A `!' after the residue type means that not all atoms were found in that sidechain, the absolute accessibilities will be correct but the meaning of the relative values is questionable, joy automatically considers these residues to be accessible. Joy currently uses a 7.0% relative total sidechain accessibility cutoff to define inaccessible residues. It is common for residues to be more than 100% accessible, this simply means that they are more accessible in the conformation they were found in, than in the reference conformation. The default is true. -w controls whether explicit water atoms are to be included in the calculations, one does not usually want to do this, thus the default is false. -h controls whether HETATM records from the input file are to be included in the surface are calculations. You must add an entry to the psa.dat file if you want to use a group that is not in the library and want to get meaningful results. (If you do not then 1.8A is assigned as the radius). The default is true. -c makes the surface calculated represent the contact surface. -s makes the calculated surface an accessible surface. In general the surface area is about three times larger than the contact area. Most programs deal with the CONTACT areas, so unless you are sure of what you are doing then leave this alone. The default is for a contact surface. FILES /usr/local/lib/psa -- default psa library directory CAVEATS The inner workings of the program are hairy, you can run into trouble with different compilers/optimization, etc. It is almost impossible to set up a new standard amino acid type (for example, a residue from an inhibitor) and get relative % accesibilty values, it is better to do this sort of calculation from the .sol file. The psa.dat file should be replaced by a set of options. SEE ALSO pdb2atm(1), joy(1) RELEASE LEVEL This document describes psa version 1.0 and later.