#Introduction

pdb_to_chains.pl is a simple script for generating a PDB file for each chain in a multi-chain PDB file. In particular, it is designed to aid in keeping a database of single-chain PDB files mirroring a local copy of the PDB.

#Usage

Usage is extremely simple. The following command will process the files 1abc.ent and 4xyz.ent:

pdb_to_chains.pl $PDBDIR 1abc.ent 4xyz.ent

It is assumed that $PDBDIR is a directory containing a copy of the RCSB's FTP archive (ftp://ftp.wwpdb.org/pub/pdb/data/structures/divided/). That is, under $PDBDIR is the following hierarchy:

$PDBDIR
└── pdb
    ├── 00
    │   ├── pdb100d.ent.gz
    │   ├── pdb200d.ent.gz
    │   ├── pdb200l.ent.gz
    │   ├── pdb300d.ent.gz
    │   └── pdb400d.ent.gz
    ├── 01
    │   ├── pdb101d.ent.gz
    │   ├── pdb101m.ent.gz
    │   ├── pdb201d.ent.gz
    │   ├── pdb201l.ent.gz
    │   ├── pdb301d.ent.gz
    │   └── pdb401d.ent.gz
    ├── 02
    │   ├── pdb102d.ent.gz
    │   ├── pdb102l.ent.gz
    │   ├── pdb102m.ent.gz
    │   ├── pdb202d.ent.gz
    SNIP

If 1abc.ent contains chains A, B and C then the files c1abcA_.pdb, c1abcB_.pdb and c1abcC_.pdb will be produced. Optionally, the -m (--middle) option may be supplied to pdb_to_chains.pl, in which case the output will be placed in subdirectories named with the middle two characters of the PDB ID (ab/c1abcA_.pdb, etc). Output is always generated in the current directory.

If the -l (--list) option is supplied, PDB files are also read from a list file.

#Rsync integration

This script is designed to work with existing cron jobs that rsync the PDB archive. As such, lines from the file specified by --list are ignored if they do not contain a file name, or if they contain the text "deleted". This allows rsync logs to be piped directly into pdb_to_chains.pl.

For example:

zcat rsyncLog.gz | pdb_to_chains.pl -l - -m