Python Script That Can Find Distance Between One Atom Of A Pdb File To All Other Atoms In Another Pdb File

Question

[Mukesh Goel]

Hi..Actually I want a script in python that can find distance between one atom of a pdb file to all other atoms in another pdb file.Otherwise I can put all the atoms in one file and find the distance between N of first residue to all N of second protein.if anyone can also guide me how to do this then also I will be highly thankful.

[Michael Schubert]

What you're asking does not make much sense. The pdb coordinates in 2 different files can be completely different (even if it's the same protein) - you would need to align your structures first in order to make any reasonable comparison.

[Philippe]

Naive question but, if you know (or verify), that the two structures are similar can't you just normalize your distance by the observed distance between a same given set of atoms (in a same plane) in the two files? I'm not used to these kind of computation so sorry if this seems stupid. =)

[Michael Schubert]

If I understand you correctly, this approach would not account for different orientation.

Answer 1

[Simon Cockell]

The distance between two atoms can be expressed as:

So your Python script needs a method that implements this for 2 sets of coordinates passed to it. eg:

import math
def distance(c1, c2):
    x_dist = (c1[0] - c2[0])**2
    y_dist = (c1[1] - c2[1])**2
    z_dist = (c1[2] - c2[2])**2
    return math.sqrt(x_dist + y_dist + z_dist)

The rest of what you want to do is just plumbing, and should be simple.

EDIT - All this said, I endorse Michael's comment above, you need to be sure the coordinates of your 2 files are actually comparable before you do any of this.

Answer 2

[brentp]

Assuming you've addressed all the issues @Michael Schubert mentions, I think you can use bio-python to do this. Something like this to find the distance from atom 111 in the B file to all atoms in the A file:

import Bio.PDB
import os.path as op
a = '1XI4.pdb'
b = '1XI4.pdb'

ATOM = 111

model_a = Bio.PDB.PDBParser().get_structure(op.splitext(a)[0], a)[0]
model_b = Bio.PDB.PDBParser().get_structure(op.splitext(b)[0], b)[0]

b_residues = list(model_b.get_residues())

print "a-index a-name distance"
for i, r in enumerate(model_a.get_residues()):
    dist = abs(r['CA'] - b_residues[ATOM]['CA'])
    print i, r.get_resname(), dist