Difference between revisions of "Ccp4 ncont"

From PyMOLWiki
Jump to: navigation, search
(updated script with faster selection and naming scheme.)
m (fixed long comment)
Line 30: Line 30:
 
def parseNCONTContacts( f ):
 
def parseNCONTContacts( f ):
 
     # /1/B/ 282(PHE). / CE1[ C]:  /1/E/ 706(GLN). / O  [ O]:  3.32
 
     # /1/B/ 282(PHE). / CE1[ C]:  /1/E/ 706(GLN). / O  [ O]:  3.32
     conParser = re.compile("\s*/(\d+)/([A-Z]*)/\s*(\d+).*?/\s*([A-Z0-9]*).*?:") # * in the second group is needed when chain code is blank
+
    # * in the second group is needed when chain code is blank
 +
     conParser = re.compile("\s*/(\d+)/([A-Z]*)/\s*(\d+).*?/\s*([A-Z0-9]*).*?:")
 
     mode = 0
 
     mode = 0
 
     s1 = []
 
     s1 = []

Revision as of 15:21, 19 June 2011

Interface residues (at cutoff <4A) in the 2c7r.pdb were found using NCONT. Usage of selectNCONTContacts script in PyMOL allows easy selection of residues and atoms listed in ncont.log file. Interacting protein and DNA residues are colored in red and slate, respectively. Atoms in contact are shown in dots.

Overview

The script selects residues and atoms from the list of the contacts found by NCONT from CCP4 Program Suite (NCONT analyses contacts between subsets of atoms in a PDB file). First, we run NCONT on our pdb file to find interface residues. Then by using the selectNCONTContacts script in PyMOL we separately select residues and atoms listed in a ncont.log file. This generates two selections (atoms and residues) for each interacting chain, allowing quick manipulation of (sometimes) extensive lists in NCONT log file.

This script works best for intermolecular contacts (when NCONT target and source selections don't overlap). If crystal contacts (NCONT parameter cell = 1 or 2) are included then additional coding is required to distinguish inter from intramolecular contacts.

Usage

selectNCONTContacts( contactsfile, selName1 = "source", selName2 = "target" )


Examples

First use NCONT to find interface residues/atoms in the pdb file. Once you have ncont.log file proceed to PyMOL. Make sure you've run the selectNCONTContacts script first.

fetch 2c7r
selectNCONTContacts ncont.log, selName1=prot, selName2=dna
Quick and easy selection of interacting residues and atoms listed in the NCONT log file. Protein and DNA residues are colored in red and slate, respectively. Atoms in contact are shown in dots.

The Code

import re

def parseNCONTContacts( f ):
    # /1/B/ 282(PHE). / CE1[ C]:  /1/E/ 706(GLN). / O  [ O]:   3.32
    # * in the second group is needed when chain code is blank
    conParser = re.compile("\s*/(\d+)/([A-Z]*)/\s*(\d+).*?/\s*([A-Z0-9]*).*?:")
    mode = 0
    s1 = []
    s2 = []
    pairs = []
    for line in f:
        if mode == 0:
            if line.strip().startswith("SOURCE ATOMS"):
                mode = 1
        elif mode == 1:
            mode = 2
        elif mode == 2:
            matches = conParser.findall(line)
            if len(matches) == 0:
                return (s1, s2, pairs)
            if len(matches) == 2:
                s1.append(matches[0])
                s2.append(matches[1])
            elif len(matches) == 1:
                s2.append(matches[0])
            pairs.append((len(s1)-1, len(s2)-1))
        else:
            print "Unknown mode", mode

def selectNCONTContacts( contactsfile, selName1 = "source", selName2 = "target" ):
    """
    selectContacts -- parses CCP4/NCONT log file and selects residues and atoms.
    http://www.ccp4.ac.uk/html/ncont.html
    
    PARAMS
        contactsfile
            filename of the CCP4/NCONT contacts log file

        selName1
            the name prefix for the _res and _atom selections returned for the
            source set of chain

        selName2
            the name prefix for the _res and _atom selections returned for the 
            target set of chain

    RETURNS
        4 selections of interface residues and atoms are created and named
        depending on what you passed into selName1 and selName2

    REPOSITORY
        https://github.com/GerhardR/pymol-scripts

    AUTHOR
        Gerhard Reitmayr and Dalia Daujotyte, 2009.
    """
    # read and parse contacts file into two lists of contact atoms and contact pair list
    s1, s2, pairs = parseNCONTContacts(open(contactsfile))
    # create a selection for the first contact list
    
    # create the PYMOL selection macros for the residues 
    resNames = [chain+"/"+residue+"/" for (type, chain, residue, atom) in s1]
    # put them in a set to remove duplicates and then join with 'or'
    resSel = " or ".join(frozenset(resNames))
    # finally select them under the new name
    cmd.select(selName1 + "_res", resSel)
    
    atomNames = [chain+"/"+residue+"/"+atom for (type, chain, residue, atom) in s1]
    atomSel = " or ".join(frozenset(atomNames))
    cmd.select(selName1 + "_atom", atomSel)

    # create a selection for the second contact list

    resNames = [chain+"/"+residue+"/" for (type, chain, residue, atom) in s2]
    resSel = " or ".join(frozenset(resNames))
    cmd.select(selName2 + "_res", resSel)
    
    atomNames = [chain+"/"+residue+"/"+atom for (type, chain, residue, atom) in s2]
    atomSel = " or ".join(frozenset(atomNames))
    cmd.select(selName2 + "_atom", atomSel)

cmd.extend("selectNCONTContacts", selectNCONTContacts)

Code repository

The latest version of this script and related scripts is available at https://github.com/GerhardR/pymol-scripts.