# Get input arguments, determine the TCR chain in use, get codon table, then load the IMGT data in
fxn.check_scripts_dir()
input_args, chain, codons = fxn.sort_input(vars(args()))
imgt_dat, tcr_functionality = fxn.get_imgt_data(chain, gene_types,
input_args['species'])
out_list, stitched = stitch(input_args, chain, imgt_dat, tcr_functionality,
codons)
out_str = '|'.join(out_list) + '(L)'
print(
'----------------------------------------------------------------------------------------------'
)
print(fxn.fastafy('nt|' + out_str, stitched))
print(fxn.fastafy('aa|' + out_str, fxn.translate_nt(stitched)))
# If a known/partial amino acid sequence provided, ensure they match up with a quick printed alignment
if 'aa' in input_args:
from Bio import pairwise2
from Bio.pairwise2 import format_alignment
alignments = pairwise2.align.globalxx(input_args['aa'],
fxn.translate_nt(stitched))
for i in range(0, 600, 60):
print('\n')
if i > len(alignments[0][0]):
break
for y in [
x[i:i + 60]
for x in format_alignment(*alignments[0]).split('\n')[:3]
preferred_alleles)
out_str = '|'.join(out_list)
# Output the appropriate strings to stdout
if input_args['mode'] not in ['BOTH_FA', 'AA_FA', 'NT_FA', 'AA', 'NT']:
raise IOError(
"Unknown output mode detected: " + input_args['mode'] + ". \n"
"Should be one of 'BOTH_FA' (default), 'AA_FA', 'NT_FA', 'AA', 'NT'."
)
if '_FA' in input_args['mode']:
print(
'----------------------------------------------------------------------------------------------'
)
if input_args['mode'] == 'BOTH_FA' or input_args['mode'] == 'NT_FA':
print(fxn.fastafy('nt|' + out_str, stitched))
if input_args['mode'] == 'BOTH_FA' or input_args['mode'] == 'AA_FA':
# Use the offset to 5' pad the stitched sequence with 'N's to make up for non-codon length 5' added seqs
print(
fxn.fastafy('aa|' + out_str,
fxn.translate_nt('N' * offset + stitched)))
elif input_args['mode'] == 'NT':
print(stitched)
elif input_args['mode'] == 'AA':
print(fxn.translate_nt('N' * offset + stitched))
# If a known/partial amino acid sequence provided, ensure they match up with a quick printed alignment
if input_args['aa']:
if __name__ == '__main__':
# Get input arguments, determine the TCR chain in use, get codon table, then load the IMGT data in
fxn.check_scripts_dir()
input_args, chain, codons = fxn.sort_input(vars(args()))
imgt_dat, tcr_functionality = fxn.get_imgt_data(chain, gene_types,
input_args['species'])
out_list, stitched = stitch(input_args, chain, imgt_dat, tcr_functionality,
codons)
out_str = '-'.join(out_list)
print '----------------------------------------------------------------------------------------------'
print fxn.fastafy('nt-' + out_str, stitched)
print fxn.fastafy('aa-' + out_str, fxn.translate_nt(stitched))
# If a known/partial amino acid sequence provided, ensure they match up with a quick printed alignment
if 'aa' in input_args:
from Bio import pairwise2
from Bio.pairwise2 import format_alignment
alignments = pairwise2.align.globalxx(input_args['aa'],
fxn.translate_nt(stitched))
for i in range(0, 600, 60):
print '\n'
if i > len(alignments[0][0]):
break
for y in [
x[i:i + 60]
for x in format_alignment(*alignments[0]).split('\n')[:3]
print("Error: imgt-data.fasta file not detected for\'", species + \
"'. Please generate and place it in the appropriate Data subdirectory.")
sys.exit()
# If so, check the modification time for the imgt-data.fasta file, assuming that's the last download time
input_imgt_file = species_dir + 'imgt-data.fasta'
mod_date = datetime.fromtimestamp(
os.path.getmtime(input_imgt_file)).strftime('%Y-%m-%d')
# Then read through the FASTA and sort into the appropriate chains
with open(input_imgt_file, 'rU') as in_file, \
open(species_dir + 'TRA.fasta', 'w') as TRA, \
open(species_dir + 'TRB.fasta', 'w') as TRB:
prot = coll.defaultdict(coll.defaultdict)
for fasta_id, seq, blank in fxn.read_fa(in_file):
gene, allele = fasta_id.split('|')[1].split('*')
# NB: TRDV included with TRA genes due to the evidence that even non 'TRAV/DV' genes can recombine with TRAJ
if 'TRA' in gene or 'TRDV' in gene:
TRA.write(fxn.fastafy(fasta_id, seq))
elif 'TRB' in gene:
TRB.write(fxn.fastafy(fasta_id, seq))
# Finally log the dates
log_txt = 'imgt-data.fasta_last_modified ' + mod_date + '\nsplit-imgt-data.py_last_run ' + fxn.today(
)
with open(species_dir + 'data-production-date.txt', 'w') as log_file:
log_file.write(log_txt)
# Can't do C checks if user providing genes, as it may be a C
if values['additional_genes'] != extra_gene_text + '\n':
tcr_bits['skip_c_checks'] = True
tcr_bits = fxn.autofill_input(tcr_bits, chain)
# Run the stitching
outputs[ref_chain + '_out_list'], \
outputs[ref_chain + '_stitched'], \
outputs[ref_chain + '_offset'] = st.stitch(tcr_bits, tcr_dat, functionality,
partial, codons, 3, preferred)
outputs[ref_chain + '_out_str'] = '|'.join(
outputs[ref_chain + '_out_list'])
outputs[ref_chain + '_fasta'] = fxn.fastafy(
'nt|' + outputs[ref_chain + '_out_str'],
outputs[ref_chain + '_stitched'])
window[ref_chain + '_out'].update(outputs[ref_chain +
'_fasta'])
except Exception as message:
warning_msgs[ref_chain + '_out'] = str(message)
elif values[ref_chain +
'V'] or values[ref_chain +
'J'] or values[ref_chain + '_CDR3']:
warnings.warn(
'V gene, J gene, and CDR3 sequence are all required to stitch a TCR chain.'
)