def align_consensi_dataset(dataset, adaIDs, fragments, VERBOSE=0):
'''Align consensi from different samples in a dataset'''
data_folder = dataset['folder']
# Collect consensi
if VERBOSE >= 1:
print 'Collecting consensi...',
consensi = defaultdict(dict)
for adaID in adaIDs:
samplename = dataset['samples'][dataset['adapters'].index(adaID)]
fragments_sample = samples[samplename]['fragments']
for frag in fragments_sample:
frag_gen = frag[:2]
if frag_gen not in fragments:
continue
con_fn = get_consensus_filename(data_folder, adaID, frag_gen)
if os.path.isfile(con_fn):
con = SeqIO.read(con_fn, 'fasta')
consensi[frag_gen][adaID] = con
if 'genomewide' in fragments:
frag_gens = [frag[:2] for frag in fragments_sample]
con_gw_fn = get_merged_consensus_filename(data_folder, adaID,
frag_gens)
if os.path.isfile(con_gw_fn):
con = SeqIO.read(con_gw_fn, 'fasta')
consensi['genomewide'][adaID] = con
if VERBOSE >= 1:
print 'done.'
print 'Aligning...',
# Align
alis = {}
for (frag, con_dict) in consensi.iteritems():
if VERBOSE >= 2:
print frag,
ali_frag = align_muscle(*(con_dict.values()))
alis[frag] = ali_frag
if VERBOSE >= 1:
print 'done.'
return alis
def align_consensi_dataset(dataset, adaIDs, fragments, VERBOSE=0):
'''Align consensi from different samples in a dataset'''
data_folder = dataset['folder']
# Collect consensi
if VERBOSE >= 1:
print 'Collecting consensi...',
consensi = defaultdict(dict)
for adaID in adaIDs:
samplename = dataset['samples'][dataset['adapters'].index(adaID)]
fragments_sample = samples[samplename]['fragments']
for frag in fragments_sample:
frag_gen = frag[:2]
if frag_gen not in fragments:
continue
con_fn = get_consensus_filename(data_folder, adaID, frag_gen)
if os.path.isfile(con_fn):
con = SeqIO.read(con_fn, 'fasta')
consensi[frag_gen][adaID] = con
if 'genomewide' in fragments:
frag_gens = [frag[:2] for frag in fragments_sample]
con_gw_fn = get_merged_consensus_filename(data_folder, adaID, frag_gens)
if os.path.isfile(con_gw_fn):
con = SeqIO.read(con_gw_fn, 'fasta')
consensi['genomewide'][adaID] = con
if VERBOSE >= 1:
print 'done.'
print 'Aligning...',
# Align
alis = {}
for (frag, con_dict) in consensi.iteritems():
if VERBOSE >= 2:
print frag,
ali_frag = align_muscle(*(con_dict.values()))
alis[frag] = ali_frag
if VERBOSE >= 1:
print 'done.'
return alis
samples = dataset.samples
if adaIDs is not None:
samples = samples.loc[samples.adapter.isin(adaIDs)]
if VERBOSE >= 3:
print 'adaIDs', samples.adapter
for samplename, sample in samples.iterrows():
sample = SampleSeq(sample)
adaID = sample.adapter
if VERBOSE >= 1:
print adaID, samplename
fragments = [fr[:2] for fr in sample.regions_complete]
if (len(fragments) != 6) and (VERBOSE >= 1):
print 'WARNING: only '+str(len(fragments))+' regions found!'
# Write one or more merged consensi
consensus = merge_consensi(data_folder, adaID, fragments, VERBOSE=VERBOSE)
for (frags, cons) in consensus:
output_filename = get_merged_consensus_filename(data_folder, adaID, frags)
SeqIO.write(cons, output_filename, 'fasta')
# Write allele frequencies
if do_nus:
nu = merge_allele_frequencies(data_folder, adaID, fragments, VERBOSE=VERBOSE)
for (frags, nuf) in nu:
output_filename = get_merged_allele_frequencies_filename(data_folder, adaID, frags)
nuf.dump(output_filename)