def get_alignment(fqs,
sample,
reference,
reference_info,
outdir,
temp,
threads,
partial=False):
paired = True if len(fqs) == 2 else False
count_file = ''.join([temp, 'pseudoalignments.tsv'])
eq_file = ''.join([temp, 'pseudoalignments.ec'])
total, avg, std = pseudoalign(fqs, sample, paired, reference, outdir, temp,
threads)
if partial:
(commithash, (gene_set, allele_idx, exon_idx, lengths, partial_exons,
partial_alleles)) = reference_info
exon_combos = get_exon_combinations()
eq_idx, align_stats = process_partial_counts(count_file, eq_file,
allele_idx, lengths,
exon_idx, exon_combos)
align_stats.extend([total, avg, std])
alignment_summary(align_stats, True)
with open(''.join([outdir, sample, '.partial_alignment.p']),
'wb') as file:
alignment_info = [commithash, eq_idx, [], paired, align_stats, []]
pickle.dump(alignment_info, file)
else:
(commithash, (gene_set, allele_idx, lengths,
gene_length)) = reference_info
eq_idx, allele_eq, align_stats = process_counts(
count_file, eq_file, gene_set, allele_idx, lengths)
align_stats.extend([total, avg, std])
alignment_summary(align_stats)
gene_stats = get_count_stats(eq_idx, gene_length)
gene_summary(gene_stats)
with open(''.join([outdir, sample, '.alignment.p']), 'wb') as file:
alignment_info = [
commithash, eq_idx, allele_eq, paired, align_stats, gene_stats
]
pickle.dump(alignment_info, file)
return alignment_info
def get_alignment(fqs,
sample,
reference,
reference_info,
outdir,
temp,
threads,
single,
partial=False,
avg=200,
std=20):
'''Runs pseudoalignment and processes output.'''
paired = not single
count_file = ''.join([temp, 'pseudoalignments.tsv'])
eq_file = ''.join([temp, 'pseudoalignments.ec'])
pseudoalign(fqs, sample, paired, reference, outdir, temp, threads, avg,
std)
# Process partial genotyping pseudoalignment
if partial:
(commithash, (gene_set, allele_idx, exon_idx, lengths, partial_exons,
partial_alleles)) = reference_info
exon_combos = get_exon_combinations()
eq_idx, align_stats = process_partial_counts(count_file, eq_file,
allele_idx, lengths,
exon_idx, exon_combos)
align_stats.extend([avg, std])
alignment_summary(align_stats, True)
with open(''.join([outdir, sample, '.partial_alignment.p']),
'wb') as file:
alignment_info = [commithash, eq_idx, [], paired, align_stats, []]
pickle.dump(alignment_info, file)
# Process regular pseudoalignment
else:
(commithash, (gene_set, allele_idx, lengths,
gene_length)) = reference_info
eq_idx, allele_eq, align_stats = process_counts(
count_file, eq_file, gene_set, allele_idx, lengths)
align_stats.extend([avg, std])
alignment_summary(align_stats)
gene_stats = get_count_stats(eq_idx, gene_length)
gene_summary(gene_stats)
with open(''.join([outdir, sample, '.alignment.p']), 'wb') as file:
alignment_info = [
commithash, eq_idx, allele_eq, paired, align_stats, gene_stats
]
pickle.dump(alignment_info, file)
with open(''.join([outdir, sample, '.genes.json']), 'w') as file:
json.dump(gene_stats, file)
return alignment_info
def get_alignment(fqs,
sample,
reference,
reference_info,
outdir,
temp,
threads,
single,
partial=False,
avg=200,
std=20):
'''Runs pseudoalignment and processes output.'''
paired = not single
count_file = ''.join([temp, 'pseudoalignments.tsv'])
eq_file = ''.join([temp, 'pseudoalignments.ec'])
total = pseudoalign(fqs, sample, paired, reference, outdir, temp, threads,
avg, std)
# Process partial genotyping pseudoalignment
if partial:
(commithash, (gene_set, allele_idx, exon_idx, lengths, partial_exons,
partial_alleles)) = reference_info
gene_set = set(gene_set)
allele_idx = json.loads(allele_idx)
exon_idx = json.loads(exon_idx)
lengths = json.loads(lengths)
lengths = dict([int(a), int(x)] for a, x in lengths.items())
partial_exons = json.loads(partial_exons)
partial_alleles = set(partial_alleles)
exon_combos = get_exon_combinations()
eq_idx, align_stats = process_partial_counts(count_file, eq_file,
allele_idx, lengths,
exon_idx, exon_combos)
align_stats.extend([total, avg, std])
alignment_summary(align_stats, True)
with open(''.join([outdir, sample, '.partial_alignment.p']),
'wb') as file:
alignment_info = [commithash, eq_idx, [], paired, align_stats, []]
pickle.dump(alignment_info, file)
# Process regular pseudoalignment
else:
(commithash, (gene_set, allele_idx, lengths,
gene_length)) = reference_info
gene_set = set(gene_set)
allele_idx = json.loads(allele_idx)
gene_length = json.loads(gene_length)
gene_length = dict([a, int(x)] for a, x in gene_length.items())
lengths = json.loads(lengths)
lengths = dict([int(a), int(x)] for a, x in lengths.items())
eq_idx, allele_eq, align_stats = process_counts(
count_file, eq_file, gene_set, allele_idx, lengths)
align_stats.extend([total, avg, std])
alignment_summary(align_stats)
gene_stats = get_count_stats(eq_idx, gene_length)
gene_summary(gene_stats)
#todo, switch to json?
with open(''.join([outdir, sample, '.alignment.p']), 'wb') as file:
alignment_info = [
commithash, eq_idx, allele_eq, paired, align_stats, gene_stats
]
pickle.dump(alignment_info, file)
with open(''.join([outdir, sample, '.genes.json']), 'w') as file:
json.dump(gene_stats, file)
return alignment_info
prior = pd.read_csv(hla_freq, delimiter='\t')
prior = prior.set_index('allele').to_dict('index')
# checks if HLA reference exists
check_ref()
# loads reference information
with open(partial_p, 'rb') as file:
(commithash, (gene_set, allele_idx, exon_idx, lengths, partial_exons,
partial_alleles)) = pickle.load(file)
log.info(f'[log] Reference: %s', commithash)
hline()
exon_combos = get_exon_combinations()
# runs transcript assembly if intermediate json not provided
reference = partial_idx
if not args.file[0].endswith('.partial_alignment.p'):
paired = True if len(args.file) == 2 else False
count_file, eq_file, num, avg, std = pseudoalign(
args.file, sample, paired, reference, outdir, temp, args.threads,
args.keep_files)
eq_idx, count = process_partial_counts(count_file, eq_file, allele_idx,
lengths, exon_idx, exon_combos,
args.keep_files)
with open(''.join([outdir, sample, '.partial_alignment.p']),