def get_other_consensi_seqrun(dataset, samplename, fragment, VERBOSE=0):
'''Get consensi of other samples except the focal one'''
consensi = []
for (samplename_other, sample) in dataset.samples.iterrows():
if samplename_other == samplename:
continue
sample = SampleSeq(sample)
ref_fn = sample.get_consensus_filename(fragment)
if not os.path.isfile(ref_fn):
if VERBOSE >= 3:
print samplename_other+': consensus for fragment '+fragment+' not found, skipping'
continue
consensi.append(SeqIO.read(ref_fn, 'fasta'))
return consensi
def get_other_consensi_seqrun(dataset, samplename, fragment, VERBOSE=0):
'''Get consensi of other samples except the focal one'''
consensi = []
for (samplename_other, sample) in dataset.samples.iterrows():
if samplename_other == samplename:
continue
sample = SampleSeq(sample)
ref_fn = sample.get_consensus_filename(fragment)
if not os.path.isfile(ref_fn):
if VERBOSE >= 3:
print samplename_other + ': consensus for fragment ' + fragment + ' not found, skipping'
continue
consensi.append(SeqIO.read(ref_fn, 'fasta'))
return consensi
def complement_consensus_PCR2(cons_rec, patient, fragment, samplen, VERBOSE=0):
'''Complement consensus from PCR2 with wings from later PCR1 sample'''
from hivwholeseq.utils.sequence import find_seed_imperfect, rfind_seed_imperfect
found = False
for _, sampletmp in patient.samples.iloc[samplen + 1:].iterrows():
for _, sampleseqtmp in sampletmp['samples seq'].iterrows():
sampleseqtmp = SampleSeq(sampleseqtmp)
if int(sampleseqtmp.PCR) == 1:
sampleseq_later = sampleseqtmp
found = True
break
if found:
break
adaID_later = sampleseq_later['adapter']
data_folder_later = sampleseq_later.sequencing_run.folder
cons_rec_later = SeqIO.read(
get_consensus_filename(data_folder_later, adaID_later, fragment),
'fasta')
conss_later = str(cons_rec_later.seq)
start = find_seed_imperfect(cons_rec_later, cons_rec[:20])
end = rfind_seed_imperfect(cons_rec_later, cons_rec[-20:]) + 20
if VERBOSE >= 1:
print 'Complementing PCR2 consensus with later PCR1:',
print sampleseq_later.name, sampleseq_later[
'seq run'], sampleseq_later.adapter
frag_spec = sampleseq_later.regions_complete[
sampleseq_later.regions_generic.index(fragment)]
return (frag_spec, conss_later[:start] + cons_rec + conss_later[end:])
# Specify the dataset
dataset = load_sequencing_run(seq_run)
data_folder = dataset.folder
# If the script is called with no adaID, iterate over all
dataset.discard_nondivided_samples()
samples = dataset.samples
if adaIDs is not None:
samples = samples.loc[samples.adapter.isin(adaIDs)]
if VERBOSE >= 3:
print 'adaIDs', samples.adapter
# Iterate over all requested samples
for samplename, sample in samples.iterrows():
sample = SampleSeq(sample)
adaID = sample.adapter
# If the script is called with no fragment, iterate over all
if not fragments:
fragments_sample = sample.regions_complete
else:
from re import findall
fragments_all = sample.regions_complete
fragments_sample = []
for fragment in fragments:
frs = filter(lambda x: fragment in x, fragments_all)
if len(frs):
fragments_sample.append(frs[0])
if 'genomewide' in fragments:
fragments_sample.append('genomewide')
if adaIDs is not None:
samples = samples.loc[samples.adapter.isin(adaIDs)]
if use_pats:
samples = samples.loc[samples['patient sample'] != 'nan']
else:
samples = samples.loc[samplenames]
if fragments is None:
fragments = ['F'+str(i+1) for i in xrange(6)]
alis = {fr: AlignIO.read(get_consensi_alignment_filename('all', fr), 'fasta')
for fr in fragments}
for samplename, sample in samples.iterrows():
sample = SampleSeq(sample)
data_folder = sample.seqrun_folder
adaID = sample.adapter
pname = sample.patientname
for fragment in fragments:
if VERBOSE >= 1:
print sample['seq run'], adaID, fragment, samplename,
# Read the summary filename of the filter_mapped, and find out whether
# there are many distant reads (a few are normal)
fn = get_filter_mapped_summary_filename(data_folder, adaID, fragment)
if os.path.isfile(fn):
found = False
with open(fn, 'r') as f:
for line in f:
samples = load_samples_sequenced()
if pnames is not None:
samples = samples.loc[samples.patient.isin(pnames)]
elif samplenames is not None:
samples = samples.loc[samples.index.isin(samplenames)]
for samplename, sample in samples.iterrows():
sample = SamplePat(sample)
if VERBOSE >= 1:
print samplename
dist_hists = []
samples_seq = sample.get_sequenced_samples()
samples_seq = samples_seq.loc[samples_seq.PCR == 1]
for samplename_seq, sample_seq in samples_seq.iterrows():
sample_seq = SampleSeq(sample_seq)
data_folder = sample_seq.seqrun_folder
adaID = sample_seq.adapter
for fragment in fragments:
try:
dist_hist = get_distance_histogram(data_folder,
adaID,
fragment,
VERBOSE=VERBOSE)
except IOError:
continue
dist_hists.append((samplename_seq, fragment, dist_hist))
dist_hists.sort(key=itemgetter(1))
samplenames_pat = samples_pat.index[ind]
samples_seq = samples_seq.loc[samples_seq['patient sample'].isin(samplenames_pat)]
else:
samples_seq = samples_seq.loc[samples_seq.index.isin(samplenames)]
if VERBOSE >= 2:
print 'samples', samples_seq.index.tolist()
# If the script is called with no fragment, iterate over all
if not fragments:
fragments = ['F'+str(i) for i in xrange(1, 7)]
if VERBOSE >= 3:
print 'fragments', fragments
for samplename, sample in samples_seq.iterrows():
sample = SampleSeq(sample)
samplename_pat = sample['patient sample']
sample_pat = samples_pat.loc[samplename_pat]
sample['patient'] = pname = sample_pat.patient
PCR = int(sample.PCR)
fragments_sample = sorted(set(sample.regions_generic) & set(fragments))
if VERBOSE:
print samplename, samplename_pat, pname, PCR
if not skip_hash:
make_output_folders(pname, samplename_pat, PCR=PCR, VERBOSE=VERBOSE)
for fragment in fragments_sample:
if VERBOSE:
data_folder = dataset.folder
samples = dataset.samples
if not fragments:
fragments = ['F' + str(i + 1) for i in xrange(6)]
matrices = {}
for fragment in fragments:
if VERBOSE:
print fragment
samples_frag = samples.loc[[
os.path.isfile(
SampleSeq(s).get_mapped_filename(fragment,
type='bam',
filtered=False))
for sn, s in samples.iterrows()
]]
n_samples = len(samples_frag)
consensi = [
SeqIO.read(SampleSeq(s).get_consensus_filename(fragment), 'fasta')
for sn, s in samples_frag.iterrows()
]
labels = [(sn, s.adapter) for sn, s in samples_frag.iterrows()]
m = np.zeros((n_samples, n_samples), int)
for si, (samplename, sample) in enumerate(samples_frag.iterrows()):
if VERBOSE == 1:
print samplename,
patient.discard_nonsequenced_samples()
mkdirs(get_initial_reference_foldername(pname))
if not fragments:
fragments = ['F' + str(i) for i in xrange(1, 7)]
if VERBOSE >= 3:
print 'fragments', fragments
if samplename is None:
sample = SamplePat(patient.samples.iloc[samplen])
else:
sample = load_sample_sequenced(samplename)
for fragment in fragments:
sample_seq = SampleSeq(sample.samples_seq.iloc[repn])
seq_run = sample_seq['seq run']
adaID = sample_seq['adapter']
dataset = sample_seq.sequencing_run
data_folder = dataset.folder
if VERBOSE:
print 'Initial sample:', sample_seq.name, sample_seq['seq run'],
print sample_seq.adapter
cons_rec = SeqIO.read(
get_consensus_filename(data_folder, adaID, fragment), 'fasta')
frag_spec = sample_seq.regions_complete[\
sample_seq.regions_generic.index(fragment)]
savefig = args.savefig
only_filt = args.only_filt
# Specify the dataset
dataset = load_sequencing_run(seq_run)
data_folder = dataset['folder']
# If the script is called with no adaID, iterate over all
if not adaIDs:
adaIDs = dataset['adapters']
if VERBOSE >= 3:
print 'adaIDs', adaIDs
# Iterate over all requested samples
for adaID in adaIDs:
# If the script is called with no fragment, iterate over all
sample = SampleSeq(dataset.samples.loc[dataset.samples.adapter == adaID].iloc[0])
samplename = sample.name
if not fragments:
fragments_sample = sample.regions_generic
else:
fragments_sample = sorted(set(fragments) & set(sample.regions_generic))
if not only_filt:
plot_minor_allele_frequency(data_folder, adaID, fragments_sample,
VERBOSE=VERBOSE, savefig=savefig)
else:
plot_minor_allele_frequency_filtered(data_folder, adaID, fragments_sample,
VERBOSE=VERBOSE, savefig=savefig)