def Renumber(self, directory):
''' Renumber structures starting at 1 '''
print('\x1b[33m[.] Renumbering structures...\x1b[0m')
current = os.getcwd()
pdbfilelist = os.listdir(directory)
os.chdir(directory)
for TheFile in tqdm.tqdm(pdbfilelist):
pdb = open(TheFile, 'r')
PDB = open(TheFile + 'X', 'w')
count = 0
num = 0
AA2 = None
for line in pdb:
count += 1
AA1 = line[23:27]
if not AA1 == AA2: num += 1
final_line =line[:7]+'{:4d}'.format(count)+line[11:17]+\
line[17:21]+'A'+'{:4d}'.format(num)+line[26:]
AA2 = AA1
PDB.write(final_line)
PDB.close()
os.remove(TheFile)
os.rename(TheFile + 'X', TheFile)
os.chdir(current)
def Database():
''' This function downloads the full PDB database and cleans it up '''
''' Out put will be a directory called PDBDatabase '''
#Collect structures
os.system('rsync -rlpt -v -z --delete --port=33444 rsync.wwpdb.org::ftp/data/structures/divided/pdb/ ./DATABASE')
current = os.getcwd()
os.mkdir('PDBDatabase')
filelist = os.listdir('DATABASE')
for directories in filelist:
files = os.listdir(current + '/DATABASE/' + directories)
for afile in files:
location = (current + '/DATABASE/' + directories + '/' + afile)
print(location)
os.rename(location , current + '/PDBDatabase/' + afile)
os.system('rm -r ./DATABASE')
#Clean Database
pdbfilelist = os.listdir('PDBDatabase')
io = Bio.PDB.PDBIO()
os.chdir('PDBDatabase')
for thefile in pdbfilelist:
try:
#Open file
TheFile = current + '/PDBDatabase/' + thefile
TheName = thefile.split('.')[0].split('pdb')[1].upper()
#Extract file
InFile = gzip.open(TheFile, 'rt')
#Separate chains and save to different files
structure = Bio.PDB.PDBParser(QUIET=True).get_structure(TheName , InFile)
count = 0
for chain in structure.get_chains():
io.set_structure(chain)
io.save(structure.get_id() + '_' + chain.get_id() + '.pdb')
print('[+] Extracted' + '\t' + thefile.upper())
os.remove(TheFile)
except:
print('[-] Failed to Extracted' + '\t' + thefile.upper())
os.remove(TheFile)
os.chdir(current)
#Remove unwanted structures
current = os.getcwd()
pdbfilelist = os.listdir('PDBDatabase')
for thefile in pdbfilelist:
TheFile = current + '/PDBDatabase/' + thefile
structure = Bio.PDB.PDBParser(QUIET=True).get_structure(TheFile.split('.')[0] , TheFile)
ppb = Bio.PDB.Polypeptide.PPBuilder()
Type = ppb.build_peptides(structure , aa_only=True)
#Delete non-protein files
if Type == []:
print('[-] NOT PROTEIN\t' , thefile)
os.remove(TheFile)
else:
#Renumber residues
pdb = open(TheFile , 'r')
PDB = open(TheFile + 'X' , 'w')
count = 0
num = 0
AA2 = None
for line in pdb:
count += 1 #Sequencially number atoms
AA1 = line[23:27] #Sequencially number residues
if not AA1 == AA2:
num += 1
final_line = line[:7] + '{:4d}'.format(count) + line[11:17] + line[17:21] + 'A' + '{:4d}'.format(num) + line[26:] #Update each line to have its atoms and residues sequencially labeled, as well as being in chain A
AA2 = AA1
PDB.write(final_line) #Write to new file called motif.pdb
PDB.close()
print('[+] GOOD\t' , thefile)
os.remove(TheFile)
os.rename(TheFile + 'X' , TheFile)