def test_simple(self, config_mock, savefig_mock):
config_mock.return_value = self.config
amino_csv = StringIO("""\
seed,region,q-cutoff,query.aa.pos,refseq.aa.pos,A,C,D,E,F,G,H,I,K,L,M,N,P,Q,R,S,T,V,W,Y,*
R1-seed,R1,15,100,1,0,5,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
R1-seed,R1,15,101,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,5,0,0,0,0
R1-seed,R1,15,102,3,0,0,0,0,0,0,0,0,0,0,0,0,0,0,5,0,0,0,0,0,0
""")
expected_scores = """\
project,region,seed,q.cut,min.coverage,which.key.pos,off.score,on.score
R1,R1,R1-seed,15,5,1,-1,1
R1-and-R2,R1,R1-seed,15,5,1,-1,1
"""
scores_csv = StringIO()
amino_csv.name = 'E1234.amino.csv'
expected_calls = [
call('E1234.R1.R1.png'),
call('E1234.R1-and-R2.R1.png')
]
coverage_plot(amino_csv,
coverage_scores_csv=scores_csv,
coverage_maps_prefix='E1234')
self.assertEqual(expected_calls, savefig_mock.mock_calls)
self.assertEqual(expected_scores, scores_csv.getvalue())
def test_simple(self, config_mock, savefig_mock):
config_mock.return_value = self.config
amino_csv = StringIO("""\
seed,region,q-cutoff,query.aa.pos,refseq.aa.pos,A,C,D,E,F,G,H,I,K,L,M,N,P,Q,R,S,T,V,W,Y,*
R1-seed,R1,15,100,1,0,5,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
R1-seed,R1,15,101,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,5,0,0,0,0
R1-seed,R1,15,102,3,0,0,0,0,0,0,0,0,0,0,0,0,0,0,5,0,0,0,0,0,0
""")
expected_scores = """\
project,region,seed,q.cut,min.coverage,which.key.pos,off.score,on.score
R1,R1,R1-seed,15,5,1,-1,1
R1-and-R2,R1,R1-seed,15,5,1,-1,1
"""
scores_csv = StringIO()
amino_csv.name = 'E1234.amino.csv'
expected_calls = [call('E1234.R1.R1.png'),
call('E1234.R1-and-R2.R1.png')]
coverage_plot(amino_csv, coverage_scores_csv=scores_csv)
self.assertEqual(expected_calls, savefig_mock.mock_calls)
self.assertEqual(expected_scores, scores_csv.getvalue())
def process_sample(sample_index, run_info, data_path, pssm):
""" Process a single sample.
:param sample_index: which sample to process from the session JSON
:param run_info: run parameters loaded from the session JSON
:param str data_path: the root folder for all BaseSpace data
:param pssm: the pssm library for running G2P analysis
"""
scratch_path = os.path.join(data_path, 'scratch')
sample_info = run_info.samples[sample_index]
sample_id = sample_info['Id']
sample_name = sample_info['Name']
sample_dir = os.path.join(data_path, 'input', 'samples', sample_id, 'Data',
'Intensities', 'BaseCalls')
if not os.path.exists(sample_dir):
sample_dir = os.path.join(data_path, 'input', 'samples', sample_id)
sample_path = None
for root, _dirs, files in os.walk(sample_dir):
sample_paths = fnmatch.filter(files, '*_R1_*')
if sample_paths:
sample_path = os.path.join(root, sample_paths[0])
break
if sample_path is None:
raise RuntimeError(
'No R1 file found for sample id {}.'.format(sample_id))
sample_path2 = sample_path.replace('_R1_', '_R2_')
if not os.path.exists(sample_path2):
raise RuntimeError('R2 file missing for sample id {}: {!r}.'.format(
sample_id, sample_path2))
logger.info('Processing sample %s (%d of %d): %s (%s).', sample_id,
sample_index + 1, len(run_info.samples), sample_name,
sample_path)
sample_out_path = create_app_result(data_path,
run_info,
sample_info,
description='Mapping results',
suffix='_QC')
sample_scratch_path = os.path.join(scratch_path, sample_name)
makedirs(sample_scratch_path)
censored_path1 = os.path.join(sample_scratch_path, 'censored1.fastq')
read_summary_path1 = os.path.join(sample_scratch_path, 'read1_summary.csv')
censor_sample(sample_path, os.path.join(scratch_path, 'bad_cycles.csv'),
censored_path1, read_summary_path1)
censored_path2 = os.path.join(sample_scratch_path, 'censored2.fastq')
read_summary_path2 = os.path.join(sample_scratch_path, 'read2_summary.csv')
censor_sample(sample_path2, os.path.join(scratch_path, 'bad_cycles.csv'),
censored_path2, read_summary_path2)
logger.info('Running prelim_map (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'prelim.csv'),
'wb') as prelim_csv:
prelim_map(censored_path1, censored_path2, prelim_csv)
logger.info('Running remap (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'prelim.csv'), 'rU') as prelim_csv, \
open(os.path.join(sample_scratch_path, 'remap.csv'), 'wb') as remap_csv, \
open(os.path.join(sample_out_path, 'remap_counts.csv'), 'wb') as counts_csv, \
open(os.path.join(sample_out_path, 'remap_conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(sample_out_path, 'unmapped1.fastq'), 'w') as unmapped1, \
open(os.path.join(sample_out_path, 'unmapped2.fastq'), 'w') as unmapped2:
remap(censored_path1,
censored_path2,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
sample_scratch_path,
nthreads=1)
logger.info('Running sam2aln (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'remap.csv'), 'rU') as remap_csv, \
open(os.path.join(sample_scratch_path, 'aligned.csv'), 'wb') as aligned_csv, \
open(os.path.join(sample_out_path, 'conseq_ins.csv'), 'wb') as insert_csv, \
open(os.path.join(sample_out_path, 'failed_read.csv'), 'wb') as failed_csv:
sam2aln(remap_csv, aligned_csv, insert_csv, failed_csv)
logger.info('Running aln2counts (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'aligned.csv'), 'rU') as aligned_csv, \
open(os.path.join(sample_out_path, 'nuc.csv'), 'wb') as nuc_csv, \
open(os.path.join(sample_out_path, 'amino.csv'), 'wb') as amino_csv, \
open(os.path.join(sample_out_path, 'coord_ins.csv'), 'wb') as coord_ins_csv, \
open(os.path.join(sample_out_path, 'conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(sample_out_path, 'failed_align.csv'), 'wb') as failed_align_csv, \
open(os.path.join(sample_out_path, 'nuc_variants.csv'), 'wb') as nuc_variants_csv, \
open(os.path.join(sample_scratch_path, 'coverage_summary.csv'), 'wb') as coverage_summary_csv:
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
nuc_variants_csv,
coverage_summary_csv=coverage_summary_csv)
logger.info('Running coverage_plots (%d of %d).', sample_index + 1,
len(run_info.samples))
coverage_path = os.path.join(sample_out_path, 'coverage')
with open(os.path.join(sample_out_path, 'amino.csv'), 'rU') as amino_csv, \
open(os.path.join(sample_out_path, 'coverage_scores.csv'), 'w') as coverage_scores_csv:
coverage_plot(amino_csv,
coverage_scores_csv,
path_prefix=coverage_path)
with open(os.path.join(sample_out_path, 'coverage_scores.csv'),
'rU') as coverage_scores_csv:
reader = csv.DictReader(coverage_scores_csv)
is_v3loop_good = False
for row in reader:
if row['region'] == 'V3LOOP':
is_v3loop_good = row['on.score'] == '4'
break
if is_v3loop_good:
logger.info('Running sam_g2p (%d of %d).', sample_index + 1,
len(run_info.samples))
g2p_path = create_app_result(data_path,
run_info,
sample_info,
description='Geno To Pheno results',
suffix='_G2P')
with open(os.path.join(sample_scratch_path, 'remap.csv'), 'rU') as remap_csv, \
open(os.path.join(sample_out_path, 'nuc.csv'), 'rU') as nuc_csv, \
open(os.path.join(g2p_path, 'g2p.csv'), 'wb') as g2p_csv, \
open(os.path.join(g2p_path, 'g2p_summary.csv'), 'wb') as g2p_summary_csv:
sam_g2p(pssm=pssm,
remap_csv=remap_csv,
nuc_csv=nuc_csv,
g2p_csv=g2p_csv,
g2p_summary_csv=g2p_summary_csv,
min_count=DEFAULT_MIN_COUNT)
def process_sample(sample_index, run_info, data_path, pssm):
""" Process a single sample.
:param sample_index: which sample to process from the session JSON
:param run_info: run parameters loaded from the session JSON
:param str data_path: the root folder for all BaseSpace data
:param pssm: the pssm library for running G2P analysis
"""
scratch_path = os.path.join(data_path, 'scratch')
sample_info = run_info.samples[sample_index]
sample_id = sample_info['Id']
sample_name = sample_info['Name']
sample_dir = os.path.join(data_path,
'input',
'samples',
sample_id,
'Data',
'Intensities',
'BaseCalls')
if not os.path.exists(sample_dir):
sample_dir = os.path.join(data_path,
'input',
'samples',
sample_id)
sample_path = None
for root, _dirs, files in os.walk(sample_dir):
sample_paths = fnmatch.filter(files, '*_R1_*')
if sample_paths:
sample_path = os.path.join(root, sample_paths[0])
break
if sample_path is None:
raise RuntimeError('No R1 file found for sample id {}.'.format(sample_id))
sample_path2 = sample_path.replace('_R1_', '_R2_')
if not os.path.exists(sample_path2):
raise RuntimeError('R2 file missing for sample id {}: {!r}.'.format(
sample_id,
sample_path2))
logger.info('Processing sample %s (%d of %d): %s (%s).',
sample_id,
sample_index+1,
len(run_info.samples),
sample_name,
sample_path)
sample_out_path = create_app_result(data_path,
run_info,
sample_info,
description='Mapping results',
suffix='_QC')
sample_scratch_path = os.path.join(scratch_path, sample_name)
makedirs(sample_scratch_path)
censored_path1 = os.path.join(sample_scratch_path, 'censored1.fastq')
read_summary_path1 = os.path.join(sample_scratch_path, 'read1_summary.csv')
censor_sample(sample_path,
os.path.join(scratch_path, 'bad_cycles.csv'),
censored_path1,
read_summary_path1)
censored_path2 = os.path.join(sample_scratch_path, 'censored2.fastq')
read_summary_path2 = os.path.join(sample_scratch_path, 'read2_summary.csv')
censor_sample(sample_path2,
os.path.join(scratch_path, 'bad_cycles.csv'),
censored_path2,
read_summary_path2)
logger.info('Running prelim_map (%d of %d).', sample_index+1, len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'prelim.csv'), 'wb') as prelim_csv:
prelim_map(censored_path1,
censored_path2,
prelim_csv)
logger.info('Running remap (%d of %d).', sample_index+1, len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'prelim.csv'), 'rU') as prelim_csv, \
open(os.path.join(sample_scratch_path, 'remap.csv'), 'wb') as remap_csv, \
open(os.path.join(sample_out_path, 'remap_counts.csv'), 'wb') as counts_csv, \
open(os.path.join(sample_out_path, 'remap_conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(sample_out_path, 'unmapped1.fastq'), 'w') as unmapped1, \
open(os.path.join(sample_out_path, 'unmapped2.fastq'), 'w') as unmapped2:
remap(censored_path1,
censored_path2,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
sample_scratch_path,
nthreads=1)
logger.info('Running sam2aln (%d of %d).', sample_index+1, len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'remap.csv'), 'rU') as remap_csv, \
open(os.path.join(sample_scratch_path, 'aligned.csv'), 'wb') as aligned_csv, \
open(os.path.join(sample_out_path, 'conseq_ins.csv'), 'wb') as insert_csv, \
open(os.path.join(sample_out_path, 'failed_read.csv'), 'wb') as failed_csv:
sam2aln(remap_csv, aligned_csv, insert_csv, failed_csv)
logger.info('Running aln2counts (%d of %d).', sample_index+1, len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'aligned.csv'), 'rU') as aligned_csv, \
open(os.path.join(sample_out_path, 'nuc.csv'), 'wb') as nuc_csv, \
open(os.path.join(sample_out_path, 'amino.csv'), 'wb') as amino_csv, \
open(os.path.join(sample_out_path, 'coord_ins.csv'), 'wb') as coord_ins_csv, \
open(os.path.join(sample_out_path, 'conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(sample_out_path, 'failed_align.csv'), 'wb') as failed_align_csv, \
open(os.path.join(sample_out_path, 'nuc_variants.csv'), 'wb') as nuc_variants_csv, \
open(os.path.join(sample_scratch_path, 'coverage_summary.csv'), 'wb') as coverage_summary_csv:
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
nuc_variants_csv,
coverage_summary_csv=coverage_summary_csv)
logger.info('Running coverage_plots (%d of %d).', sample_index+1, len(run_info.samples))
coverage_path = os.path.join(sample_out_path, 'coverage')
with open(os.path.join(sample_out_path, 'amino.csv'), 'rU') as amino_csv, \
open(os.path.join(sample_out_path, 'coverage_scores.csv'), 'w') as coverage_scores_csv:
coverage_plot(amino_csv, coverage_scores_csv, path_prefix=coverage_path)
with open(os.path.join(sample_out_path, 'coverage_scores.csv'), 'rU') as coverage_scores_csv:
reader = csv.DictReader(coverage_scores_csv)
is_v3loop_good = False
for row in reader:
if row['region'] == 'V3LOOP':
is_v3loop_good = row['on.score'] == '4'
break
if is_v3loop_good:
logger.info('Running sam_g2p (%d of %d).', sample_index+1, len(run_info.samples))
g2p_path = create_app_result(data_path,
run_info,
sample_info,
description='Geno To Pheno results',
suffix='_G2P')
with open(os.path.join(sample_scratch_path, 'remap.csv'), 'rU') as remap_csv, \
open(os.path.join(sample_out_path, 'nuc.csv'), 'rU') as nuc_csv, \
open(os.path.join(g2p_path, 'g2p.csv'), 'wb') as g2p_csv, \
open(os.path.join(g2p_path, 'g2p_summary.csv'), 'wb') as g2p_summary_csv:
sam_g2p(pssm=pssm,
remap_csv=remap_csv,
nuc_csv=nuc_csv,
g2p_csv=g2p_csv,
g2p_summary_csv=g2p_summary_csv,
min_count=DEFAULT_MIN_COUNT)
def process_sample(self, fastq1, progress, prefixes, image_paths,
error_log):
fastq2 = fastq1.replace('_R1_001',
'_R2_001').replace('censored1', 'censored2')
if not os.path.exists(fastq2):
raise IOError('ERROR: Missing R2 file for {}'.format(fastq1))
prefix = os.path.basename(fastq1).replace('_L001_R1_001.fastq',
'').replace(
'.censored1.fastq', '')
prefixes.append(prefix)
output_csv = prefix + '.prelim.csv'
self.write('Processing sample {} ({})\n'.format(prefix, progress))
with open(output_csv, 'wb') as handle:
prelim_map(fastq1,
fastq2,
handle,
nthreads=self.nthreads,
callback=self.callback,
stderr=error_log)
# prepare file handles for remap stage
with open(output_csv, 'rU') as prelim_csv, \
open(os.path.join(self.workdir, prefix + '.remap.csv'), 'wb') as remap_csv, \
open(os.path.join(self.workdir, prefix + '.remap_counts.csv'), 'wb') as counts_csv, \
open(os.path.join(self.workdir, prefix + '.remap_conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(self.workdir, prefix + '.unmapped1.fastq'), 'w') as unmapped1, \
open(os.path.join(self.workdir, prefix + '.unmapped2.fastq'), 'w') as unmapped2:
self.write('... remapping\n')
self.parent.update()
self.progress_bar['value'] = 0
remap(fastq1,
fastq2,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
self.workdir,
nthreads=self.nthreads,
callback=self.callback,
stderr=error_log)
# prepare file handles for conversion from SAM format to alignment
with open(os.path.join(self.workdir, prefix + '.remap.csv'), 'rU') as remap_csv, \
open(os.path.join(self.workdir, prefix + '.aligned.csv'), 'wb') as aligned_csv, \
open(os.path.join(self.workdir, prefix + '.insert.csv'), 'wb') as insert_csv, \
open(os.path.join(self.workdir, prefix + '.failed.csv'), 'wb') as failed_csv:
self.write('... converting into alignment\n')
self.parent.update()
sam2aln(remap_csv,
aligned_csv,
insert_csv,
failed_csv,
nthreads=self.nthreads)
with open(os.path.join(self.workdir, prefix + '.aligned.csv'), 'rU') as aligned_csv, \
open(os.path.join(self.workdir, prefix + '.nuc.csv'), 'wb') as nuc_csv, \
open(os.path.join(self.workdir, prefix + '.amino.csv'), 'wb') as amino_csv, \
open(os.path.join(self.workdir, prefix + '.coord_ins.csv'), 'wb') as coord_ins_csv, \
open(os.path.join(self.workdir, prefix + '.conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(self.workdir, prefix + '.failed_align.csv'), 'wb') as failed_align_csv, \
open(os.path.join(self.workdir, prefix + '.nuc_variants.csv'), 'wb') as nuc_variants_csv:
self.parent.update()
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
nuc_variants_csv,
callback=self.callback)
self.write('... generating coverage plots\n')
self.parent.update()
with open(os.path.join(self.workdir, prefix + '.amino.csv'),
'rU') as amino_csv:
image_paths += coverage_plot(amino_csv)
self.write('... performing g2p scoring on samples covering HIV-1 V3\n')
self.parent.update()
with open(os.path.join(self.workdir, prefix + '.remap.csv'), 'rU') as remap_csv, \
open(os.path.join(self.workdir, prefix + '.nuc.csv'), 'rU') as nuc_csv, \
open(os.path.join(self.workdir, prefix + '.g2p.csv'), 'wb') as g2p_csv:
sam_g2p(pssm=self.pssm,
remap_csv=remap_csv,
nuc_csv=nuc_csv,
g2p_csv=g2p_csv)
def process_sample(self, fastq1, progress, prefixes, image_paths, error_log):
fastq2 = fastq1.replace('_R1_001', '_R2_001').replace('censored1',
'censored2')
if not os.path.exists(fastq2):
raise IOError('ERROR: Missing R2 file for {}'.format(fastq1))
prefix = os.path.basename(fastq1).replace('_L001_R1_001.fastq',
'').replace('.censored1.fastq',
'')
prefixes.append(prefix)
output_csv = prefix + '.prelim.csv'
self.write('Processing sample {} ({})\n'.format(prefix, progress))
with open(output_csv, 'wb') as handle:
prelim_map(fastq1,
fastq2,
handle,
nthreads=self.nthreads,
callback=self.callback,
stderr=error_log)
# prepare file handles for remap stage
with open(output_csv, 'rU') as prelim_csv, \
open(os.path.join(self.workdir, prefix + '.remap.csv'), 'wb') as remap_csv, \
open(os.path.join(self.workdir, prefix + '.remap_counts.csv'), 'wb') as counts_csv, \
open(os.path.join(self.workdir, prefix + '.remap_conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(self.workdir, prefix + '.unmapped1.fastq'), 'w') as unmapped1, \
open(os.path.join(self.workdir, prefix + '.unmapped2.fastq'), 'w') as unmapped2:
self.write('... remapping\n')
self.parent.update()
self.progress_bar['value'] = 0
remap(fastq1,
fastq2,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
self.workdir,
nthreads=self.nthreads,
callback=self.callback,
stderr=error_log)
# prepare file handles for conversion from SAM format to alignment
with open(os.path.join(self.workdir, prefix + '.remap.csv'), 'rU') as remap_csv, \
open(os.path.join(self.workdir, prefix + '.aligned.csv'), 'wb') as aligned_csv, \
open(os.path.join(self.workdir, prefix + '.insert.csv'), 'wb') as insert_csv, \
open(os.path.join(self.workdir, prefix + '.failed.csv'), 'wb') as failed_csv:
self.write('... converting into alignment\n')
self.parent.update()
sam2aln(remap_csv,
aligned_csv,
insert_csv,
failed_csv,
nthreads=self.nthreads)
with open(os.path.join(self.workdir, prefix + '.aligned.csv'), 'rU') as aligned_csv, \
open(os.path.join(self.workdir, prefix + '.nuc.csv'), 'wb') as nuc_csv, \
open(os.path.join(self.workdir, prefix + '.amino.csv'), 'wb') as amino_csv, \
open(os.path.join(self.workdir, prefix + '.coord_ins.csv'), 'wb') as coord_ins_csv, \
open(os.path.join(self.workdir, prefix + '.conseq.csv'), 'wb') as conseq_csv, \
open(os.path.join(self.workdir, prefix + '.failed_align.csv'), 'wb') as failed_align_csv, \
open(os.path.join(self.workdir, prefix + '.nuc_variants.csv'), 'wb') as nuc_variants_csv:
self.parent.update()
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
nuc_variants_csv,
callback=self.callback)
self.write('... generating coverage plots\n')
self.parent.update()
with open(os.path.join(self.workdir, prefix + '.amino.csv'), 'rU') as amino_csv:
image_paths += coverage_plot(amino_csv)
self.write('... performing g2p scoring on samples covering HIV-1 V3\n')
self.parent.update()
with open(os.path.join(self.workdir, prefix + '.remap.csv'), 'rU') as remap_csv, \
open(os.path.join(self.workdir, prefix + '.nuc.csv'), 'rU') as nuc_csv, \
open(os.path.join(self.workdir, prefix + '.g2p.csv'), 'wb') as g2p_csv:
sam_g2p(pssm=self.pssm,
remap_csv=remap_csv,
nuc_csv=nuc_csv,
g2p_csv=g2p_csv)
