def download_pdbs(base_dir, protein_codes):
"""
Downloads the PDB database (or a part of it) as PDB files. Every protein is stored in it's own
directory (with name the PDB code) under base_dir.
:param base_dir: where to download all the proteins.
:param protein_codes: the PDB codes of the proteins that should be downloaded.
"""
prot_codes = []
if isinstance(protein_codes, dict):
for key in protein_codes.keys():
prot_codes += protein_codes[key]
else:
prot_codes = protein_codes
prot_codes = list(set(prot_codes))
from Bio.PDB import PDBList
failed = 0
attempted = len(prot_codes)
for code in prot_codes:
try:
pl = PDBList(pdb=os.path.join(base_dir, code.upper()))
pl.flat_tree = 1
pl.retrieve_pdb_file(pdb_code=code)
except IOError:
log.warning("Failed to download protein {}".format(code))
failed += 1
continue
log.info("Downloaded {0}/{1} molecules".format(attempted - failed,
attempted))
from configure import configure
config = configure()
from Bio.PDB import PDBList
from os.path import dirname, join, basename
from glob import glob
from os import remove
pdb_dir = dirname(config["cif"])
pl = PDBList(pdb=pdb_dir)
pl.flat_tree = True
existingCifs = glob(join(pdb_dir, "*.cif"))
existingPdbs = set([])
for cif in existingCifs:
existingPdbs.add(basename(cif).replace(".cif", ""))
allPdbs = set(pdb_code.lower() for pdb_code in pl.get_all_entries())
pdb2delete = existingPdbs - allPdbs
pdb2download = allPdbs - existingPdbs
print("Found ", len(pdb2delete), " files to delete")
print("and ", len(pdb2download), " file to download")
for pdb_code in pdb2download:
pl.retrieve_pdb_file(pdb_code, file_format="mmCif")
for pdb_code in pdb2delete:
cifPath = join(pdb_dir, pdb_code + ".cif")
remove(cifPath)
