Splitting several PDB files into chains and save as separate files

Question

[Jonathan Lefebre]

I am trying to split a large number of pdb files using Biopython and then save them as separate files called pdbid_chain.pdb . So far I did not succeed. Additionally, I am quite new to python.

Any help is highly appreciated!

Here is my code:

#pdb_list contains a list of 208 pdb structures 
#PDB_RAW_DIR is the directory where structures are stored
io = PDBIO()

#parse structures
for f in pdb_list:
    pdb_parsed = PDBParser().get_structure(pdb_ids, str(PDB_RAW_DIR) + '/' + f)

#save chains
for structure in pdb_parsed:
        pdb_chains = structure.get_chains()
        for chain in pdb_chains:
            io.set_structure(chain)
            io.save(pdb_parsed.get_id() + "_" + chain.get_id() + ".pdb")

Cheers!

Answer 1

[Mensur Dlakic]

It is not that difficult to figure out BioPython's I/O manipulations with PDB files - see here.

I will get you started with a basic code, and you should be able to figure out how to work with a list. Let's download the files:

wget -q -o /dev/null ftp://ftp.ebi.ac.uk/pub/databases/msd/pdb_uncompressed/pdb2f2f.ent
mv pdb2f2f.ent 2f2f.pdb

The code is hard-coded to work with the file above, but with small changes it can be made more universal.

from Bio.PDB import PDBParser
from Bio.PDB.PDBIO import PDBIO

parser = PDBParser()
io = PDBIO()

structure = parser.get_structure("2f2f", "2f2f.pdb")
pdb_chains = structure.get_chains()
for chain in pdb_chains:
    io.set_structure(chain)
    io.save(structure.get_id() + "_" + chain.get_id() + ".pdb")

This will crate 6 different files for chains A-F.

[Wayne]

Alternatives:

You can view a static version of a notebook here that has a bash/sed version of splitting out the chains that is very short.
At the top of that notebook you can find a link to use and run an R-based version, pdbsplit, in the Bio3D package.
You can run that notebooks actively by going here, clicking the launch binder badge, and then in the session that comes up choose the from the available notebooks, 'Split PDB files into chains using command line'.

[Mensur Dlakic]

There is also a stand-alone program called pdbsplitchains in this library.

[Jonathan Lefebre]

Thanks a lot Mensur! But this does not really help as this is a solution for a single file and is basically what I already did. Meanwhile, I figure out that it is a fundamental problem with biopython (I am using v. 1.77), as described here: https://github.com/biopython/biopython/pull/3223 The fix that they are proposing is working.

Thanks anyways!