def _test(self, read_indexes, debug_file_prefix=None):
read_indexes = reversed(read_indexes)
simple_filename1 = self.filename1 + '_simple.fastq'
self.write_simple_fastq(simple_filename1, read_indexes)
workdir = os.path.dirname(self.filename1)
os.chdir(workdir)
simple_filename2 = get_reverse_filename(simple_filename1)
censored_filename1 = os.path.join(workdir, 'rerun.censored1.fastq')
censored_filename2 = os.path.join(workdir, 'rerun.censored2.fastq')
trimmed_filename1 = os.path.join(workdir, 'rerun.trimmed1.fastq')
trimmed_filename2 = os.path.join(workdir, 'rerun.trimmed2.fastq')
prelim_censored_filename = os.path.join(workdir, 'rerun_censored.prelim.csv')
prelim_trimmed_filename = os.path.join(workdir, 'rerun_trimmed.prelim.csv')
with open(self.bad_cycles_filename, 'rU') as bad_cycles:
bad_cycles = list(csv.DictReader(bad_cycles))
with open(simple_filename1, 'rU') as simple1, \
open(censored_filename1, 'w') as censored1:
censor(simple1, bad_cycles, censored1, use_gzip=False)
with open(simple_filename2, 'rU') as simple2, \
open(censored_filename2, 'w') as censored2:
censor(simple2, bad_cycles, censored2, use_gzip=False)
with open(prelim_censored_filename, 'w+') as prelim_censored_csv, \
open(prelim_trimmed_filename, 'w+') as prelim_trimmed_csv:
prelim_map(censored_filename1,
censored_filename2,
prelim_censored_csv,
nthreads=BOWTIE_THREADS)
trim((simple_filename1, simple_filename2),
self.bad_cycles_filename,
(trimmed_filename1, trimmed_filename2),
use_gzip=False)
prelim_map(trimmed_filename1,
trimmed_filename2,
prelim_trimmed_csv,
nthreads=BOWTIE_THREADS)
prelim_censored_csv.seek(0)
censored_map_count = self.count_mapped(prelim_censored_csv)
prelim_trimmed_csv.seek(0)
trimmed_map_count = self.count_mapped(prelim_trimmed_csv)
return self.get_result(censored_map_count, trimmed_map_count)
def _test(self, read_indexes, debug_file_prefix=None):
read_indexes = reversed(read_indexes)
simple_filename1 = self.filename1 + '_simple.fastq'
self.write_simple_fastq(simple_filename1, read_indexes)
workdir = os.path.dirname(self.filename1)
os.chdir(workdir)
simple_filename2 = get_reverse_filename(simple_filename1)
censored_filename1 = os.path.join(workdir, 'rerun.censored1.fastq')
censored_filename2 = os.path.join(workdir, 'rerun.censored2.fastq')
trimmed_filename1 = os.path.join(workdir, 'rerun.trimmed1.fastq')
trimmed_filename2 = os.path.join(workdir, 'rerun.trimmed2.fastq')
prelim_censored_filename = os.path.join(workdir, 'rerun_censored.prelim.csv')
prelim_trimmed_filename = os.path.join(workdir, 'rerun_trimmed.prelim.csv')
with open(self.bad_cycles_filename, 'rU') as bad_cycles:
bad_cycles = list(csv.DictReader(bad_cycles))
with open(simple_filename1, 'rU') as simple1, \
open(censored_filename1, 'w') as censored1:
censor(simple1, bad_cycles, censored1, use_gzip=False)
with open(simple_filename2, 'rU') as simple2, \
open(censored_filename2, 'w') as censored2:
censor(simple2, bad_cycles, censored2, use_gzip=False)
with open(prelim_censored_filename, 'w+') as prelim_censored_csv, \
open(prelim_trimmed_filename, 'w+') as prelim_trimmed_csv:
prelim_map(censored_filename1,
censored_filename2,
prelim_censored_csv,
nthreads=BOWTIE_THREADS)
trim((simple_filename1, simple_filename2),
self.bad_cycles_filename,
(trimmed_filename1, trimmed_filename2),
use_gzip=False)
prelim_map(trimmed_filename1,
trimmed_filename2,
prelim_trimmed_csv,
nthreads=BOWTIE_THREADS)
prelim_censored_csv.seek(0)
censored_map_count = self.count_mapped(prelim_censored_csv)
prelim_trimmed_csv.seek(0)
trimmed_map_count = self.count_mapped(prelim_trimmed_csv)
return self.get_result(censored_map_count, trimmed_map_count)
def test_trim(tmpdir):
read1_content = 'TATCTACTAACTGTCGGTCTAC'
read2_content = reverse_and_complement(read1_content)
expected1 = build_fastq(read1_content)
expected2 = build_fastq(read2_content)
tmp_path = Path(str(tmpdir))
fastq1_path = tmp_path / 'read1.fastq'
fastq2_path = tmp_path / 'read2.fastq'
trimmed1_path = tmp_path / 'trimmed1.fastq'
trimmed2_path = tmp_path / 'trimmed2.fastq'
fastq1_path.write_text(expected1)
fastq2_path.write_text(expected2)
trim([fastq1_path, fastq2_path],
'no_bad_cycles.csv',
[str(trimmed1_path), str(trimmed2_path)],
use_gzip=False)
trimmed1 = trimmed1_path.read_text()
trimmed2 = trimmed2_path.read_text()
assert trimmed1 == expected1
assert trimmed2 == expected2
def process(self,
pssm,
excluded_seeds=(),
excluded_projects=(),
force_gzip=False,
use_denovo=False):
""" Process a single sample.
:param pssm: the pssm library for running G2P analysis
:param excluded_seeds: seeds to exclude from mapping
:param excluded_projects: project codes to exclude from reporting
:param bool force_gzip: treat FASTQ files as gzipped, even when they
don't end in .gz
:param bool use_denovo: True if de novo assembly should be used,
instead of bowtie2 mapping against references.
"""
logger.info('Processing %s (%r).', self, self.fastq1)
scratch_path = self.get_scratch_path()
makedirs(scratch_path)
use_gzip = force_gzip or self.fastq1.endswith('.gz')
sample_info = self.load_sample_info()
with open(self.read_summary_csv, 'w') as read_summary:
trim((self.fastq1, self.fastq2),
self.bad_cycles_csv,
(self.trimmed1_fastq, self.trimmed2_fastq),
summary_file=read_summary,
use_gzip=use_gzip,
skip=self.skip,
project_code=sample_info.get('project'))
if use_denovo:
logger.info('Running merge_for_entropy on %s.', self)
with open(self.read_entropy_csv, 'w') as read_entropy_csv:
merge_for_entropy(self.trimmed1_fastq,
self.trimmed2_fastq,
read_entropy_csv,
scratch_path)
write_merge_lengths_plot(self.read_entropy_csv,
self.merge_lengths_svg)
logger.info('Running fastq_g2p on %s.', self)
with open(self.trimmed1_fastq) as fastq1, \
open(self.trimmed2_fastq) as fastq2, \
open(self.g2p_csv, 'w') as g2p_csv, \
open(self.g2p_summary_csv, 'w') as g2p_summary_csv, \
open(self.g2p_unmapped1_fastq, 'w') as g2p_unmapped1, \
open(self.g2p_unmapped2_fastq, 'w') as g2p_unmapped2, \
open(self.g2p_aligned_csv, 'w') as g2p_aligned_csv, \
open(self.merged_contigs_csv, 'w') as merged_contigs_csv:
fastq_g2p(pssm=pssm,
fastq1=fastq1,
fastq2=fastq2,
g2p_csv=g2p_csv,
g2p_summary_csv=g2p_summary_csv,
unmapped1=g2p_unmapped1,
unmapped2=g2p_unmapped2,
aligned_csv=g2p_aligned_csv,
min_count=DEFAULT_MIN_COUNT,
min_valid=MIN_VALID,
min_valid_percent=MIN_VALID_PERCENT,
merged_contigs_csv=merged_contigs_csv)
if use_denovo:
self.run_denovo(excluded_seeds)
else:
self.run_mapping(excluded_seeds)
logger.info('Running sam2aln on %s.', self)
with open(self.remap_csv) as remap_csv, \
open(self.aligned_csv, 'w') as aligned_csv, \
open(self.conseq_ins_csv, 'w') as conseq_ins_csv, \
open(self.failed_csv, 'w') as failed_csv, \
open(self.clipping_csv, 'w') as clipping_csv:
sam2aln(remap_csv,
aligned_csv,
conseq_ins_csv,
failed_csv,
clipping_csv=clipping_csv)
logger.info('Running aln2counts on %s.', self)
if use_denovo:
contigs_path = self.contigs_csv
else:
contigs_path = os.devnull
with open(self.aligned_csv) as aligned_csv, \
open(self.g2p_aligned_csv) as g2p_aligned_csv, \
open(self.clipping_csv) as clipping_csv, \
open(self.conseq_ins_csv) as conseq_ins_csv, \
open(self.remap_conseq_csv) as remap_conseq_csv, \
open(contigs_path) as contigs_csv, \
open(self.nuc_csv, 'w') as nuc_csv, \
open(self.nuc_detail_csv, 'w') as nuc_detail_csv, \
open(self.amino_csv, 'w') as amino_csv, \
open(self.amino_detail_csv, 'w') as amino_detail_csv, \
open(self.coord_ins_csv, 'w') as coord_ins_csv, \
open(self.conseq_csv, 'w') as conseq_csv, \
open(self.conseq_region_csv, 'w') as conseq_region_csv, \
open(self.failed_align_csv, 'w') as failed_align_csv, \
open(self.coverage_summary_csv, 'w') as coverage_summary_csv, \
open(self.genome_coverage_csv, 'w') as genome_coverage_csv, \
open(self.conseq_all_csv, "w") as conseq_all_csv, \
open(self.minimap_hits_csv, "w") as minimap_hits_csv:
if not use_denovo:
for f in (amino_detail_csv, nuc_detail_csv):
f.close()
os.remove(f.name)
amino_detail_csv = nuc_detail_csv = None
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
coverage_summary_csv=coverage_summary_csv,
clipping_csv=clipping_csv,
conseq_ins_csv=conseq_ins_csv,
g2p_aligned_csv=g2p_aligned_csv,
remap_conseq_csv=remap_conseq_csv,
conseq_region_csv=conseq_region_csv,
amino_detail_csv=amino_detail_csv,
nuc_detail_csv=nuc_detail_csv,
genome_coverage_csv=genome_coverage_csv,
contigs_csv=contigs_csv,
conseq_all_csv=conseq_all_csv,
minimap_hits_csv=minimap_hits_csv)
logger.info('Running coverage_plots on %s.', self)
os.makedirs(self.coverage_maps)
with open(self.amino_csv) as amino_csv, \
open(self.coverage_scores_csv, 'w') as coverage_scores_csv:
coverage_plot(amino_csv,
coverage_scores_csv,
coverage_maps_path=self.coverage_maps,
coverage_maps_prefix=self.name,
excluded_projects=excluded_projects)
with open(self.genome_coverage_csv) as genome_coverage_csv, \
open(self.minimap_hits_csv) as minimap_hits_csv:
if not use_denovo:
minimap_hits_csv = None
plot_genome_coverage(genome_coverage_csv,
minimap_hits_csv,
self.genome_coverage_svg)
logger.info('Running cascade_report on %s.', self)
with open(self.g2p_summary_csv) as g2p_summary_csv, \
open(self.remap_counts_csv) as remap_counts_csv, \
open(self.aligned_csv) as aligned_csv, \
open(self.cascade_csv, 'w') as cascade_csv:
cascade_report = CascadeReport(cascade_csv)
cascade_report.g2p_summary_csv = g2p_summary_csv
cascade_report.remap_counts_csv = remap_counts_csv
cascade_report.aligned_csv = aligned_csv
cascade_report.generate()
logger.info('Finished sample %s.', self)
def process(self,
pssm,
excluded_seeds=(),
excluded_projects=(),
force_gzip=False):
""" Process a single sample.
:param pssm: the pssm library for running G2P analysis
:param excluded_seeds: seeds to exclude from mapping
:param excluded_projects: project codes to exclude from reporting
:param bool force_gzip: treat FASTQ files as gzipped, even when they
don't end in .gz
"""
logger.info('Processing %s (%r).', self, self.fastq1)
scratch_path = os.path.dirname(self.prelim_csv)
os.mkdir(scratch_path)
use_gzip = force_gzip or self.fastq1.endswith('.gz')
with open(self.read_summary_csv, 'w') as read_summary:
trim((self.fastq1, self.fastq2),
self.bad_cycles_csv,
(self.trimmed1_fastq, self.trimmed2_fastq),
summary_file=read_summary,
use_gzip=use_gzip)
logger.info('Running fastq_g2p on %s.', self)
with open(self.trimmed1_fastq) as fastq1, \
open(self.trimmed2_fastq) as fastq2, \
open(self.g2p_csv, 'w') as g2p_csv, \
open(self.g2p_summary_csv, 'w') as g2p_summary_csv, \
open(self.g2p_unmapped1_fastq, 'w') as g2p_unmapped1, \
open(self.g2p_unmapped2_fastq, 'w') as g2p_unmapped2, \
open(self.g2p_aligned_csv, 'w') as g2p_aligned_csv:
fastq_g2p(pssm=pssm,
fastq1=fastq1,
fastq2=fastq2,
g2p_csv=g2p_csv,
g2p_summary_csv=g2p_summary_csv,
unmapped1=g2p_unmapped1,
unmapped2=g2p_unmapped2,
aligned_csv=g2p_aligned_csv,
min_count=DEFAULT_MIN_COUNT,
min_valid=MIN_VALID,
min_valid_percent=MIN_VALID_PERCENT)
logger.info('Running prelim_map on %s.', self)
with open(self.prelim_csv, 'w') as prelim_csv:
prelim_map(self.g2p_unmapped1_fastq,
self.g2p_unmapped2_fastq,
prelim_csv,
work_path=scratch_path,
excluded_seeds=excluded_seeds)
logger.info('Running remap on %s.', self)
if self.debug_remap:
debug_file_prefix = os.path.join(scratch_path, 'debug')
else:
debug_file_prefix = None
with open(self.prelim_csv) as prelim_csv, \
open(self.remap_csv, 'w') as remap_csv, \
open(self.remap_counts_csv, 'w') as counts_csv, \
open(self.remap_conseq_csv, 'w') as conseq_csv, \
open(self.unmapped1_fastq, 'w') as unmapped1, \
open(self.unmapped2_fastq, 'w') as unmapped2:
remap(self.g2p_unmapped1_fastq,
self.g2p_unmapped2_fastq,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
scratch_path,
debug_file_prefix=debug_file_prefix)
logger.info('Running sam2aln on %s.', self)
with open(self.remap_csv) as remap_csv, \
open(self.aligned_csv, 'w') as aligned_csv, \
open(self.conseq_ins_csv, 'w') as conseq_ins_csv, \
open(self.failed_csv, 'w') as failed_csv, \
open(self.clipping_csv, 'w') as clipping_csv:
sam2aln(remap_csv,
aligned_csv,
conseq_ins_csv,
failed_csv,
clipping_csv=clipping_csv)
logger.info('Running aln2counts on %s.', self)
with open(self.aligned_csv) as aligned_csv, \
open(self.g2p_aligned_csv) as g2p_aligned_csv, \
open(self.clipping_csv) as clipping_csv, \
open(self.conseq_ins_csv) as conseq_ins_csv, \
open(self.remap_conseq_csv) as remap_conseq_csv, \
open(self.nuc_csv, 'w') as nuc_csv, \
open(self.amino_csv, 'w') as amino_csv, \
open(self.coord_ins_csv, 'w') as coord_ins_csv, \
open(self.conseq_csv, 'w') as conseq_csv, \
open(self.conseq_region_csv, 'w') as conseq_region_csv, \
open(self.failed_align_csv, 'w') as failed_align_csv, \
open(self.coverage_summary_csv, 'w') as coverage_summary_csv:
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
coverage_summary_csv=coverage_summary_csv,
clipping_csv=clipping_csv,
conseq_ins_csv=conseq_ins_csv,
g2p_aligned_csv=g2p_aligned_csv,
remap_conseq_csv=remap_conseq_csv,
conseq_region_csv=conseq_region_csv)
logger.info('Running coverage_plots on %s.', self)
os.makedirs(self.coverage_maps)
with open(self.amino_csv) as amino_csv, \
open(self.coverage_scores_csv, 'w') as coverage_scores_csv:
coverage_plot(amino_csv,
coverage_scores_csv,
coverage_maps_path=self.coverage_maps,
coverage_maps_prefix=self.name,
excluded_projects=excluded_projects)
logger.info('Running cascade_report on %s.', self)
with open(self.g2p_summary_csv) as g2p_summary_csv, \
open(self.remap_counts_csv) as remap_counts_csv, \
open(self.aligned_csv) as aligned_csv, \
open(self.cascade_csv, 'w') as cascade_csv:
cascade_report = CascadeReport(cascade_csv)
cascade_report.g2p_summary_csv = g2p_summary_csv
cascade_report.remap_counts_csv = remap_counts_csv
cascade_report.aligned_csv = aligned_csv
cascade_report.generate()
logger.info('Finished sample %s.', self)
def process_sample(sample_index, run_info, args, 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 args: the command-line arguments
:param pssm: the pssm library for running G2P analysis
"""
scratch_path = os.path.join(args.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(args.data_path, 'input', 'samples', sample_id,
'Data', 'Intensities', 'BaseCalls')
if not os.path.exists(sample_dir):
sample_dir = os.path.join(args.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_qc_path = os.path.join(args.qc_path, sample_name)
makedirs(sample_qc_path)
sample_scratch_path = os.path.join(scratch_path, sample_name)
makedirs(sample_scratch_path)
bad_cycles_path = os.path.join(scratch_path, 'bad_cycles.csv')
trimmed_path1 = os.path.join(sample_scratch_path, 'trimmed1.fastq')
read_summary_path = os.path.join(sample_scratch_path, 'read_summary.csv')
trimmed_path2 = os.path.join(sample_scratch_path, 'trimmed2.fastq')
with open(read_summary_path, 'w') as read_summary:
trim((sample_path, sample_path2),
bad_cycles_path, (trimmed_path1, trimmed_path2),
summary_file=read_summary,
use_gzip=sample_path.endswith('.gz'))
logger.info('Running fastq_g2p (%d of %d).', sample_index + 1,
len(run_info.samples))
g2p_unmapped1_path = os.path.join(sample_scratch_path,
'g2p_unmapped1.fastq')
g2p_unmapped2_path = os.path.join(sample_scratch_path,
'g2p_unmapped2.fastq')
with open(os.path.join(sample_scratch_path, 'trimmed1.fastq'), 'r') as fastq1, \
open(os.path.join(sample_scratch_path, 'trimmed2.fastq'), 'r') as fastq2, \
open(os.path.join(sample_scratch_path, 'g2p.csv'), 'w') as g2p_csv, \
open(os.path.join(sample_scratch_path, 'g2p_summary.csv'), 'w') as g2p_summary_csv, \
open(g2p_unmapped1_path, 'w') as g2p_unmapped1, \
open(g2p_unmapped2_path, 'w') as g2p_unmapped2, \
open(os.path.join(sample_scratch_path, 'g2p_aligned.csv'), 'w') as g2p_aligned_csv:
fastq_g2p(pssm=pssm,
fastq1=fastq1,
fastq2=fastq2,
g2p_csv=g2p_csv,
g2p_summary_csv=g2p_summary_csv,
unmapped1=g2p_unmapped1,
unmapped2=g2p_unmapped2,
aligned_csv=g2p_aligned_csv,
min_count=DEFAULT_MIN_COUNT,
min_valid=MIN_VALID,
min_valid_percent=MIN_VALID_PERCENT)
logger.info('Running prelim_map (%d of %d).', sample_index + 1,
len(run_info.samples))
excluded_seeds = [] if args.all_projects else EXCLUDED_SEEDS
with open(os.path.join(sample_scratch_path, 'prelim.csv'),
'w') as prelim_csv:
prelim_map(g2p_unmapped1_path,
g2p_unmapped2_path,
prelim_csv,
work_path=sample_scratch_path,
excluded_seeds=excluded_seeds)
logger.info('Running remap (%d of %d).', sample_index + 1,
len(run_info.samples))
if args.debug_remap:
debug_file_prefix = os.path.join(sample_scratch_path, 'debug')
else:
debug_file_prefix = None
with open(os.path.join(sample_scratch_path, 'prelim.csv'), 'r') as prelim_csv, \
open(os.path.join(sample_scratch_path, 'remap.csv'), 'w') as remap_csv, \
open(os.path.join(sample_scratch_path, 'remap_counts.csv'), 'w') as counts_csv, \
open(os.path.join(sample_scratch_path, 'remap_conseq.csv'), 'w') as conseq_csv, \
open(os.path.join(sample_qc_path, 'unmapped1.fastq'), 'w') as unmapped1, \
open(os.path.join(sample_qc_path, 'unmapped2.fastq'), 'w') as unmapped2:
remap(g2p_unmapped1_path,
g2p_unmapped2_path,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
sample_scratch_path,
debug_file_prefix=debug_file_prefix)
logger.info('Running sam2aln (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'remap.csv'), 'r') as remap_csv, \
open(os.path.join(sample_scratch_path, 'aligned.csv'), 'w') as aligned_csv, \
open(os.path.join(sample_scratch_path, 'conseq_ins.csv'), 'w') as conseq_ins_csv, \
open(os.path.join(sample_scratch_path, 'failed_read.csv'), 'w') as failed_csv, \
open(os.path.join(sample_scratch_path, 'clipping.csv'), 'w') as clipping_csv:
sam2aln(remap_csv,
aligned_csv,
conseq_ins_csv,
failed_csv,
clipping_csv=clipping_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'), 'r') as aligned_csv, \
open(os.path.join(sample_scratch_path, 'g2p_aligned.csv'), 'r') as g2p_aligned_csv, \
open(os.path.join(sample_scratch_path, 'clipping.csv'), 'r') as clipping_csv, \
open(os.path.join(sample_scratch_path, 'conseq_ins.csv'), 'r') as conseq_ins_csv, \
open(os.path.join(sample_scratch_path, 'remap_conseq.csv'), 'r') as remap_conseq_csv, \
open(os.path.join(sample_scratch_path, 'nuc.csv'), 'w') as nuc_csv, \
open(os.path.join(sample_scratch_path, 'amino.csv'), 'w') as amino_csv, \
open(os.path.join(sample_scratch_path, 'coord_ins.csv'), 'w') as coord_ins_csv, \
open(os.path.join(sample_scratch_path, 'conseq.csv'), 'w') as conseq_csv, \
open(os.path.join(sample_scratch_path, 'failed_align.csv'), 'w') as failed_align_csv, \
open(os.path.join(sample_scratch_path, 'coverage_summary.csv'), 'w') as coverage_summary_csv:
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
coverage_summary_csv=coverage_summary_csv,
clipping_csv=clipping_csv,
conseq_ins_csv=conseq_ins_csv,
g2p_aligned_csv=g2p_aligned_csv,
remap_conseq_csv=remap_conseq_csv)
logger.info('Running coverage_plots (%d of %d).', sample_index + 1,
len(run_info.samples))
coverage_maps_path = os.path.join(args.qc_path, 'coverage_maps')
makedirs(coverage_maps_path)
excluded_projects = [] if args.all_projects else EXCLUDED_PROJECTS
with open(os.path.join(sample_scratch_path, 'amino.csv'), 'r') as amino_csv, \
open(os.path.join(sample_scratch_path, 'coverage_scores.csv'), 'w') as coverage_scores_csv:
coverage_plot(amino_csv,
coverage_scores_csv,
coverage_maps_path=coverage_maps_path,
coverage_maps_prefix=sample_name,
excluded_projects=excluded_projects)
logger.info('Running hivdb (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'amino.csv')) as amino_csv, \
open(os.path.join(sample_scratch_path, 'coverage_scores.csv')) as coverage_scores_csv, \
open(os.path.join(sample_scratch_path, 'resistance.csv'), 'w') as resistance_csv, \
open(os.path.join(sample_scratch_path, 'mutations.csv'), 'w') as mutations_csv:
hivdb(amino_csv, coverage_scores_csv, resistance_csv, mutations_csv,
run_info.reports)
logger.info('Running resistance report (%d of %d).', sample_index + 1,
len(run_info.samples))
source_path = os.path.dirname(__file__)
version_filename = os.path.join(source_path, 'version.txt')
if not os.path.exists(version_filename):
git_version = 'v0-dev'
else:
with open(version_filename) as version_file:
git_version = version_file.read().strip()
reports_path = os.path.join(args.qc_path, 'resistance_reports')
makedirs(reports_path)
report_filename = os.path.join(reports_path,
sample_name + '_resistance.pdf')
with open(os.path.join(sample_scratch_path, 'resistance.csv')) as resistance_csv, \
open(os.path.join(sample_scratch_path, 'mutations.csv')) as mutations_csv, \
open(report_filename, 'wb') as report_pdf:
gen_report(resistance_csv,
mutations_csv,
report_pdf,
sample_name,
git_version=git_version)
logger.info('Running cascade_report (%d of %d).', sample_index + 1,
len(run_info.samples))
with open(os.path.join(sample_scratch_path, 'g2p_summary.csv'), 'r') as g2p_summary_csv, \
open(os.path.join(sample_scratch_path, 'remap_counts.csv'), 'r') as remap_counts_csv, \
open(os.path.join(sample_scratch_path, 'aligned.csv'), 'r') as aligned_csv, \
open(os.path.join(sample_scratch_path, 'cascade.csv'), 'w') as cascade_csv:
cascade_report = CascadeReport(cascade_csv)
cascade_report.g2p_summary_csv = g2p_summary_csv
cascade_report.remap_counts_csv = remap_counts_csv
cascade_report.aligned_csv = aligned_csv
cascade_report.generate()
logger.info('Finished sample (%d of %d).', sample_index + 1,
len(run_info.samples))
def process(self,
pssm,
excluded_seeds=(),
excluded_projects=(),
force_gzip=False):
""" Process a single sample.
:param pssm: the pssm library for running G2P analysis
:param excluded_seeds: seeds to exclude from mapping
:param excluded_projects: project codes to exclude from reporting
:param bool force_gzip: treat FASTQ files as gzipped, even when they
don't end in .gz
"""
logger.info('Processing %s (%r).', self, self.fastq1)
scratch_path = os.path.dirname(self.prelim_csv)
os.mkdir(scratch_path)
use_gzip = force_gzip or self.fastq1.endswith('.gz')
with open(self.read_summary_csv, 'w') as read_summary:
trim((self.fastq1, self.fastq2),
self.bad_cycles_csv,
(self.trimmed1_fastq, self.trimmed2_fastq),
summary_file=read_summary,
use_gzip=use_gzip)
logger.info('Running fastq_g2p on %s.', self)
with open(self.trimmed1_fastq) as fastq1, \
open(self.trimmed2_fastq) as fastq2, \
open(self.g2p_csv, 'w') as g2p_csv, \
open(self.g2p_summary_csv, 'w') as g2p_summary_csv, \
open(self.g2p_unmapped1_fastq, 'w') as g2p_unmapped1, \
open(self.g2p_unmapped2_fastq, 'w') as g2p_unmapped2, \
open(self.g2p_aligned_csv, 'w') as g2p_aligned_csv:
fastq_g2p(pssm=pssm,
fastq1=fastq1,
fastq2=fastq2,
g2p_csv=g2p_csv,
g2p_summary_csv=g2p_summary_csv,
unmapped1=g2p_unmapped1,
unmapped2=g2p_unmapped2,
aligned_csv=g2p_aligned_csv,
min_count=DEFAULT_MIN_COUNT,
min_valid=MIN_VALID,
min_valid_percent=MIN_VALID_PERCENT)
logger.info('Running prelim_map on %s.', self)
with open(self.prelim_csv, 'w') as prelim_csv:
prelim_map(self.g2p_unmapped1_fastq,
self.g2p_unmapped2_fastq,
prelim_csv,
work_path=scratch_path,
excluded_seeds=excluded_seeds)
logger.info('Running remap on %s.', self)
if self.debug_remap:
debug_file_prefix = os.path.join(scratch_path, 'debug')
else:
debug_file_prefix = None
with open(self.prelim_csv) as prelim_csv, \
open(self.remap_csv, 'w') as remap_csv, \
open(self.remap_counts_csv, 'w') as counts_csv, \
open(self.remap_conseq_csv, 'w') as conseq_csv, \
open(self.unmapped1_fastq, 'w') as unmapped1, \
open(self.unmapped2_fastq, 'w') as unmapped2:
remap(self.g2p_unmapped1_fastq,
self.g2p_unmapped2_fastq,
prelim_csv,
remap_csv,
counts_csv,
conseq_csv,
unmapped1,
unmapped2,
scratch_path,
debug_file_prefix=debug_file_prefix)
logger.info('Running sam2aln on %s.', self)
with open(self.remap_csv) as remap_csv, \
open(self.aligned_csv, 'w') as aligned_csv, \
open(self.conseq_ins_csv, 'w') as conseq_ins_csv, \
open(self.failed_csv, 'w') as failed_csv, \
open(self.clipping_csv, 'w') as clipping_csv:
sam2aln(remap_csv,
aligned_csv,
conseq_ins_csv,
failed_csv,
clipping_csv=clipping_csv)
logger.info('Running aln2counts on %s.', self)
with open(self.aligned_csv) as aligned_csv, \
open(self.g2p_aligned_csv) as g2p_aligned_csv, \
open(self.clipping_csv) as clipping_csv, \
open(self.conseq_ins_csv) as conseq_ins_csv, \
open(self.remap_conseq_csv) as remap_conseq_csv, \
open(self.nuc_csv, 'w') as nuc_csv, \
open(self.amino_csv, 'w') as amino_csv, \
open(self.coord_ins_csv, 'w') as coord_ins_csv, \
open(self.conseq_csv, 'w') as conseq_csv, \
open(self.conseq_region_csv, 'w') as conseq_region_csv, \
open(self.failed_align_csv, 'w') as failed_align_csv, \
open(self.coverage_summary_csv, 'w') as coverage_summary_csv:
aln2counts(aligned_csv,
nuc_csv,
amino_csv,
coord_ins_csv,
conseq_csv,
failed_align_csv,
coverage_summary_csv=coverage_summary_csv,
clipping_csv=clipping_csv,
conseq_ins_csv=conseq_ins_csv,
g2p_aligned_csv=g2p_aligned_csv,
remap_conseq_csv=remap_conseq_csv,
conseq_region_csv=conseq_region_csv)
logger.info('Running coverage_plots on %s.', self)
os.makedirs(self.coverage_maps)
with open(self.amino_csv) as amino_csv, \
open(self.coverage_scores_csv, 'w') as coverage_scores_csv:
coverage_plot(amino_csv,
coverage_scores_csv,
coverage_maps_path=self.coverage_maps,
coverage_maps_prefix=self.name,
excluded_projects=excluded_projects)
logger.info('Running cascade_report on %s.', self)
with open(self.g2p_summary_csv) as g2p_summary_csv, \
open(self.remap_counts_csv) as remap_counts_csv, \
open(self.aligned_csv) as aligned_csv, \
open(self.cascade_csv, 'w') as cascade_csv:
cascade_report = CascadeReport(cascade_csv)
cascade_report.g2p_summary_csv = g2p_summary_csv
cascade_report.remap_counts_csv = remap_counts_csv
cascade_report.aligned_csv = aligned_csv
cascade_report.generate()
logger.info('Finished sample %s.', self)