def compute_dists(index):
ntcr = sample_tcrs.loc[index, ] #tcrr
edists = []
for tindex, tcr in epigroup.iterrows():
if tcr['tcr_info'] == ntcr['tcr_info']:
continue
edists.append(
ch_tcr_distances.compute_distance(tcr['tcr_info'],
ntcr['tcr_info'],
chains,
distance_params,
rep_dists=rep_dists))
edists.sort()
for nbrdist_percentile in nbrdist_percentiles:
nbrdist = ch_tcr_distances.sort_and_compute_nbrdist_from_distances(
edists, nbrdist_percentile, dont_sort=True)
wtd_nbrdist = ch_tcr_distances.sort_and_compute_weighted_nbrdist_from_distances(
edists, nbrdist_percentile, dont_sort=True)
ntcr['{}_{}_nbrdist{}'.format(
epitope, chains,
nbrdist_percentile)] = '{:.3f}'.format(nbrdist)
ntcr['{}_{}_wtd_nbrdist{}'.format(
epitope, chains,
nbrdist_percentile)] = '{:.3f}'.format(wtd_nbrdist)
return ntcr
def compute_dists(index):
ntcr = all_tcrs.loc[index, ]
edists = []
for tindex, tcr in epigroup.iterrows():
if tcr['tcr_info'] == ntcr['tcr_info'] or tindex == index:
#if epitope=='LLWNGPMAV':
# print(tcr['tcr_info'], ntcr['tcr_info'], 'continued')
continue
edists.append(
ch_tcr_distances.compute_distance(tcr['tcr_info'],
ntcr['tcr_info'],
chains,
distance_params,
rep_dists=rep_dists))
if edists[-1] == 0:
print(tcr['tcr_info'], ntcr['tcr_info'], 'added to dists')
edists.sort()
for nbrdist_percentile in nbrdist_percentiles:
nbrdist = ch_tcr_distances.sort_and_compute_nbrdist_from_distances(
edists, nbrdist_percentile, dont_sort=True)
wtd_nbrdist = ch_tcr_distances.sort_and_compute_weighted_nbrdist_from_distances(
edists, nbrdist_percentile, dont_sort=True)
ntcr['{}_{}_nbrdist{}'.format(
epitope, chains, nbrdist_percentile)] = '{:.3f}'.format(nbrdist)
ntcr['{}_{}_wtd_nbrdist{}'.format(
epitope, chains,
nbrdist_percentile)] = '{:.3f}'.format(wtd_nbrdist)
return ntcr
def compute_dists(rindex):
random_tcr = random_tcrs.loc[rindex, ]
edists = []
for tindex, tcr in epigroup.iterrows():
edists.append(ch_tcr_distances.compute_distance(tcr['tcr_info'], random_tcr['tcr_info'],
chains, distance_params, rep_dists=rep_dists))
edists.sort()
for nbrdist_percentile in nbrdist_percentiles:
nbrdist = ch_tcr_distances.sort_and_compute_nbrdist_from_distances(edists, nbrdist_percentile,
dont_sort=True)
wtd_nbrdist = ch_tcr_distances.sort_and_compute_weighted_nbrdist_from_distances(edists,
nbrdist_percentile,
dont_sort=True)
random_tcr['{}_{}_nbrdist{}'.format(epitope, chains, nbrdist_percentile)] = '{:.3f}'.format(
nbrdist)
random_tcr['{}_{}_wtd_nbrdist{}'.format(epitope, chains, nbrdist_percentile)] = '{:.3f}'.format(
wtd_nbrdist)
return random_tcr
def compute_dists(rindex):
random_tcr = random_tcrs.loc[rindex, ]
edists = []
for tindex, tcr in epigroup.iterrows():
edists.append(ch_tcr_distances.compute_distance(tcr['tcr_info'], random_tcr['tcr_info'],
chains, distance_params, rep_dists=rep_dists))
edists.sort()
for nbrdist_percentile in nbrdist_percentiles:
nbrdist = ch_tcr_distances.sort_and_compute_nbrdist_from_distances(edists, nbrdist_percentile,
dont_sort=True)
wtd_nbrdist = ch_tcr_distances.sort_and_compute_weighted_nbrdist_from_distances(edists,
nbrdist_percentile,
dont_sort=True)
random_tcr['{}_{}_nbrdist{}'.format(epitope, chains, nbrdist_percentile)] = '{:.3f}'.format(
nbrdist)
random_tcr['{}_{}_wtd_nbrdist{}'.format(epitope, chains, nbrdist_percentile)] = '{:.3f}'.format(
wtd_nbrdist)
return random_tcr
rep_dists = ch_tcr_distances.compute_all_v_region_distances_shugay(vmatrix, organism, distance_params)
else:
rep_dists = ch_tcr_distances.compute_all_v_region_distances(organism, distance_params)
tcr_col = ['cdr1.alpha', 'cdr2.alpha', 'cdr2.5.alpha', 'cdr3.alpha',
'cdr1.beta', 'cdr2.beta', 'cdr2.5.beta', 'cdr3.beta',
'v.alpha', 'v.beta', epitope_col, 'species']
all_tcrs = ch_read_files.read_tcr(clones_file, organism, chains, epitope_col, tcr_col)
tcr_col_end = ['cdr3.1', 'v.1', '{}.1'.format(epitope_col),
'cdr3.2', 'v.2', '{}.2'.format(epitope_col),
'distance']
#sys.exit()
with open(outfile, 'w') as out:
out.write('{}\n'.format('\t'.join(tcr_col_end)))
for iti, tcr in all_tcrs.iterrows():
print(iti)
for _, ntcr in all_tcrs.iterrows():
distance = ch_tcr_distances.compute_distance(tcr['tcr_info'], ntcr['tcr_info'],
chains, distance_params, rep_dists=rep_dists)
out.write('{}\t{}\t{}\t{}\t{}\t{}\t{}\n'.format(tcr['cdr3.beta'], tcr['v.beta'], tcr[epitope_col],
ntcr['cdr3.beta'], ntcr['v.beta'], ntcr[epitope_col],
distance))
print('Computation of distances is completed')