if __name__ == '__main__':
# Get input arguments, get the required data read in
fxn.check_scripts_dir()
input_args = vars(args())
codons = fxn.get_optimal_codons(input_args['codon_usage'],
input_args['species'].upper())
linker_dict = fxn.get_linker_dict()
tcr_dat = {}
tcr_functionality = {}
for c in ['TRA', 'TRB']:
tmp_tcr_dat, tmp_functionality = fxn.get_imgt_data(
c, st.gene_types, input_args['species'].upper())
tcr_dat[c] = tmp_tcr_dat
tcr_functionality[c] = tmp_functionality
# Then go through in file and stitch each TCR on each line
if not os.path.isfile(input_args['in_file']):
raise IOError(
input_args['in_file'] +
" not detected - please check and specify in file again.")
# TODO opener function for gzipped
with open(input_args['in_file'], 'rU') as in_file:
line_count = 0
out_data = ['\t'.join(out_headers)]
regions = {
'v': 'V-REGION',
'j': 'J-REGION',
'c': 'EX1+EX2+EX3+EX4',
'l': 'L-PART1+L-PART2'
}
gene_types = list(regions.values())
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) + '(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
# Then stitch each individual chain...
for ref_chain in ['TR1', 'TR2']:
chain = convert_chains[receptor][ref_chain]
window[ref_chain + '_out'].update('')
window[ref_chain + '_log'].update('')
with warnings.catch_warnings(record=True) as chain_log:
warnings.simplefilter("always")
if values[ref_chain +
'V'] and values[ref_chain +
'J'] and values[ref_chain + '_CDR3']:
try:
tcr_dat, functionality, partial = fxn.get_imgt_data(
chain, st.gene_types, species)
# If additional genes provided, just add them to all possible gene segment types
if values['additional_genes'] != extra_gene_text + '\n':
for extra_gene in outputs['additional_fastas']:
gene, allele = extra_gene[0].split('*')
for gene_type in tcr_dat.keys():
if gene not in tcr_dat[gene_type]:
tcr_dat[gene_type][
gene] = coll.defaultdict(list)
if allele in tcr_dat[gene_type][gene]:
raise warnings.warn(
"User provided gene/allele combination "
else: # If not explicitly provided, infer from input TSV headers
with fxn.opener(input_args['in_file']) as in_file:
for line in in_file:
if 'TRAV' in line and 'TRGV' not in line:
receptor = 'TRA/TRB'
elif 'TRGV' in line and 'TRAV' not in line:
receptor = 'TRG/TRD'
else:
raise IOError("Unable to determine receptor from input file header, please check template. ")
break
# Define the individual receptors (chains or loci, i.e. TRA and TRB or TRG and TRD) in play
r1, r2 = receptor.split('/')
for c in [r1, r2]:
tmp_tcr_dat, tmp_functionality, partial = fxn.get_imgt_data(c, st.gene_types, species)
tcr_dat[c] = tmp_tcr_dat
tcr_functionality[c] = tmp_functionality
if 'extra_genes' in input_args:
if input_args['extra_genes']:
tcr_dat[c], tcr_functionality[c] = fxn.get_additional_genes(tcr_dat[c], tcr_functionality[c])
input_args['skip_c_checks'] = True
else:
input_args['skip_c_checks'] = False
# Allow for provision of preferred alleles
if input_args['preferred_alleles_path']:
preferences[c] = fxn.get_preferred_alleles(input_args['preferred_alleles_path'], list(fxn.regions.values()),
tcr_dat[c], partial, c)
else:
if __name__ == '__main__':
# TODO move all this to one large bracketing function?
# 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()))
regions = {
'v': 'V-REGION',
'j': 'J-REGION',
'c': 'EX1+EX2+EX3+EX4',
'l': 'L-PART1+L-PART2'
}
gene_types = regions.values()
imgt_dat, functionality = fxn.get_imgt_data(chain, gene_types)
# Then find each of the appropriate sequences
done = {}
for r in regions:
if '*' in input_args[r]:
gene, allele = input_args[r].split('*')
if allele not in imgt_dat[regions[r]][gene]:
print "\tCannot find", r.upper(), "gene", input_args[r] + \
": attempting prototypical allele (" + gene + "*01)"
allele = '01'
else:
gene = input_args[r]
allele = '01'