How can we find the distance between all residues in a PDB file?

Question

If we have a PDB structrure, how can we find residues physically interacting with each other in space? I know that we must find the distance between residues and if the distance is less than 5-6 Angstrom, we say that residues are physically interacting. But how can we find the distance between all residues and how can we finally determine the distances between all residues? Is there a software or webserver for that?

Answer 1

If you need to process multiple files, you could use Biopython to parse a PDB structure.

from Bio.PDB import PDBParser

# create parser
parser = PDBParser()

# read structure from file
structure = parser.get_structure('PHA-L', '1fat.pdb')

model = structure[0]
chain = model['A']

# this example uses only the first residue of a single chain.
# it is easy to extend this to multiple chains and residues.
for residue1 in chain:
    for residue2 in chain:
        if residue1 != residue2:
            # compute distance between CA atoms
            try:
                distance = residue1['CA'] - residue2['CA']
            except KeyError:
                ## no CA atom, e.g. for H_NAG
                continue
            if distance < 6:
                print(residue1, residue2, distance)
        # stop after first residue
        break

If you need to look at one structure, using a viewer perhaps would be easier. You could try PyMOL: (how to measure distance). There are other PDB viewers, some of which can work even through a browser.

Answer 2

Could you use CCP4's NCONT program? There's a GUI and a command line interface, whatever suits. You can specify which chains you want to target and interact with and set a cut off for distance. The bonus here is once you're in you have a nice suite of other structural tools to use.

If you're just doing it once, the GUI is friendly enough to work things out, if you're doing a batch then you can run it across several files via the command line.

Download

Answer 3

As part of a project me and some teammates did a script that outputs visual maps of distances between residues. It uses Biopython.

The module contact_map.py does what you are looking for. As an example, if you want to find the residues whose CA are below 5 you can run the following command:

python3 contact_map.py pdb1cd8.ent -a CA -CA 5 

This will produce three files:

distance_map_pdb1cd8_CA.png # Heatmap of the distance between the residues
contact_map_pdb1cd8_CA.png # Black/White heatmap: If it is at that min distance
contact_map.log  # The actions taken

If you don't have downloaded already the pdb structue you can use the main module:

python3 cozmic.py real 1cd8 -a CA -CA 5 

Answer 4

You can use MDtraj. The package is easy to install using Anaconda.

You can get the interacting residues using the following snippet (taken from http://mdtraj.org/1.6.2/examples/native-contact.html)

heavy_pairs = np.array(
    [(i,j) for (i,j) in combinations(heavy, 2)
        if abs(native.topology.atom(i).residue.index - \
               native.topology.atom(j).residue.index) > 3])

# compute the distances between these pairs in the native state
heavy_pairs_distances = md.compute_distances(native[0], heavy_pairs)[0]
# and get the pairs s.t. the distance is less than NATIVE_CUTOFF
native_contacts = heavy_pairs[heavy_pairs_distances < NATIVE_CUTOFF]
print("Number of native contacts", len(native_contacts))

Answer 5

Distance between C-alphas can also be found using the BioStructures.jl package in Julia:

using BioStructures
struc = read("1AKE.pdb", PDB)
calphas = collectatoms(struc['A'], calphaselector)
dm = DistanceMap(calphas)

Then dm[5, 10] gives 12.19677604943208 etc.

Other variations also work, see the docs.