generate_gene_modles.py is a Python script for rapidly mining the Protein Data Bank and generating representative structural models of a given human gene. The script first downloads all PDB entries associated with the gene (using the accompanying script generate_pdb_models.py). Then the script compares all PDB structures and selects representative structures optimizing for sequence coverage, structural resolution, and structural diversity (i.e. multiple conformations).
usage: generate_gene_models.py [-h] gene dest
Mine the PDB for structural models of a gene
positional arguments:
gene UNIPROT primary gene name
dest Destination parent directory for the gene
optional arguments:
-h, --help show this help message and exit
gene - The Uniprot Primary gene name for the human gene to generate representative structural models of
dest - The top-level directory in which to store the structural models of gene (will be created if it doesn't already exist)
The output model files are stored in the 'dest' directory and are organized by PDB and then by chain. The naming system is . Assume you run the following command:
python generate_gene_models.py geneX ~/geneX
The directory ~/geneX would be organized as follows:
~/geneX/
representative_gene_models/ * This directory contains the representative gene models
geneX_pdbID_chainID_iso/
geneX_pdbID_chainID_iso.pdb * This is a representative gene model. It is chain <chainID> of the structure of geneX with PDB ID <pdbID>
... * There may be multiple representative structural model of geneX
geneX_pdbID/ * The rest of the directories at this level contain structures specific to different PDB entries for geneX
representative_pdb_models/ * This directory contains the representative structures (chains) extracted from PDB <pdbID>
geneX_pdbID_chainID_iso/
geneX_pdbID_chainID_iso.pdb * This is a representative pdb model. It is chain <chainID> of the structure of geneX with PDB ID <pdbID>
geneX_pdbID.pdb * The original PDB file
geneX_pdbID_chainID/
geneX_pdbID_chainID.pdb * This is chain <chainID> from PDB <pdbID>
... * There may be multiple unique chains in <pdbID> that are all geneX
geneX_pdbID_molinfo.xml * metadata from the PDB
geneX_pdbID_pdbinfo.xml * metadata from the PDB
... * There may be multiple pdbIDs corresponding to geneX