def genreport_rerun(source, working):
run_path = os.path.dirname(os.path.dirname(source))
run_name = os.path.basename(run_path)
publish_path = os.path.join(working, 'rerun_results', run_name)
os.makedirs(publish_path)
print('##', run_name)
working_paths = split_files(source, working)
file_names = (os.path.basename(working_path)
for working_path in working_paths)
sorted_file_names = sorted(file_names)
sample_sheet_path = os.path.join(source, '../../SampleSheet.csv')
groups = list(
find_groups(sorted_file_names,
sample_sheet_path,
included_projects={'HCV'}))
for group in groups:
working_path = os.path.join(working, group.names[0])
if group.names[1] is None:
midi_name = ''
midi_path = working_path
else:
midi_name = group.names[1]
midi_path = os.path.join(working, group.names[1])
if not os.path.isdir(midi_path):
midi_name = 'failed MidHCV'
midi_path = working_path
print(working_path, midi_name)
with open(os.path.join(working_path, 'amino.csv')) as amino_csv, \
open(os.path.join(midi_path, 'amino.csv')) as midi_amino_csv, \
open(os.path.join(working_path, 'resistance.csv'), 'w') as resistance_csv, \
open(os.path.join(working_path, 'mutations.csv'), 'w') as mutations_csv, \
open(os.path.join(working_path, 'resistance_fail.csv'), 'w') as resistance_fail_csv:
report_resistance(amino_csv, midi_amino_csv, resistance_csv,
mutations_csv, resistance_fail_csv)
sample_name = os.path.basename(working_path)
with open(os.path.join(working_path, 'resistance.csv')) as resistance_csv, \
open(os.path.join(working_path, 'mutations.csv')) as mutations_csv, \
open(os.path.join(working_path, 'resistance_report.pdf'), 'wb') as resistance_report_csv:
gen_report(resistance_csv,
mutations_csv,
resistance_report_csv,
sample_name=sample_name)
for file_name in ('resistance.csv', 'mutations.csv',
'resistance_fail.csv'):
with open(os.path.join(publish_path, file_name), 'w') as dest:
dest_writer = csv.writer(dest)
for i, group in enumerate(groups):
working_path = os.path.join(working, group.names[0])
sample_name = os.path.basename(working_path)
with open(os.path.join(working_path, file_name),
'r') as source:
source_reader = csv.reader(source)
for j, row in enumerate(source_reader):
if j != 0:
row.insert(0, sample_name)
elif i == 0:
row.insert(0, 'sample')
else:
continue
dest_writer.writerow(row)
def test_find_groups_checks_overrides(tmpdir):
sample_sheet_path = Path(str(tmpdir)) / 'SampleSheet.csv'
sample_sheet_path.write_text(BASIC_HEADER + """\
Sample_ID,Sample_Name,Sample_Plate,Sample_Well,index,index2,Sample_Project,Description,GenomeFolder
CFE_SomeId_10-Jul-2014_N501-N701_Sample1_Proj1,Sample1_Proj1,,,ACGTACGT,TGCATGCA,,,
CFE_SomeId_10-Jul-2014_N501-N702_Sample2_Proj2,Sample2_Proj2,,,AAAAGGGG,CCCCTTTT,,,
""")
sample_sheet_overrides_path = sample_sheet_path.parent / 'SampleSheetOverrides.csv'
sample_sheet_overrides_path.write_text("""\
sample,project
Sample1-Proj1_S1,AltProjA
""")
expected_groups = [
SampleGroup('Sample1', ('Sample1-Proj1_S1_L001_R1_001.fastq.gz', None),
('AltProjA', None)),
SampleGroup('Sample2', ('Sample2-Proj2_S2_L001_R1_001.fastq.gz', None),
('Proj2', None))
]
groups = list(
find_groups([
'Sample1-Proj1_S1_L001_R1_001.fastq.gz',
'Sample2-Proj2_S2_L001_R1_001.fastq.gz'
], sample_sheet_path))
assert expected_groups == groups
def link_samples(run_path, data_path):
""" Load the data from a run folder into the BaseSpace layout. """
shutil.rmtree(data_path, ignore_errors=True)
makedirs(data_path)
results_path = os.path.join(run_path, 'Results', 'basespace')
makedirs(results_path)
output_path = os.path.join(data_path, 'output')
os.symlink(results_path, output_path)
scratch_path = os.path.join(data_path, 'scratch')
makedirs(scratch_path)
sample_groups = []
run_info_path = os.path.join(run_path, 'RunInfo.xml')
interop_path = os.path.join(run_path, 'InterOp')
if not (os.path.exists(run_info_path) and os.path.exists(interop_path)):
read_sizes = None
else:
read_sizes = parse_read_sizes(run_info_path)
run_info = RunInfo(sample_groups,
reports=['PR_RT', 'IN', 'NS3', 'NS5a', 'NS5b'],
interop_path=interop_path,
scratch_path=scratch_path,
output_path=output_path,
read_sizes=read_sizes)
fastq_files = list(glob(os.path.join(run_path,
'Data',
'Intensities',
'BaseCalls',
'*_R1_*')) or
glob(os.path.join(run_path,
'*_R1_*')))
source_folder = fastq_files and os.path.dirname(fastq_files[0])
file_names = [os.path.basename(fastq_file) for fastq_file in fastq_files]
groups = find_groups(file_names,
os.path.join(run_path, 'SampleSheet.csv'))
for group in groups:
main_file, midi_file = group.names
if main_file.startswith('Undetermined'):
continue
main_sample = Sample(fastq1=os.path.join(source_folder, main_file))
if midi_file is None:
midi_sample = None
else:
midi_sample = Sample(fastq1=os.path.join(source_folder, midi_file))
sample_groups.append(SampleGroup(main_sample, midi_sample))
sample_count = sum(1 for _ in run_info.get_all_samples())
for i, sample in enumerate(run_info.get_all_samples(), 1):
sample.rank = '({} of {})'.format(i, sample_count)
sample.bad_cycles_csv = run_info.bad_cycles_csv
sample.scratch_path = os.path.join(scratch_path, sample.name)
return run_info
def find_full_groups(fastq_files, sandbox_path):
groups = list(
find_groups([p.name for p in fastq_files],
sandbox_path / 'SampleSheet.csv'))
full_groups = []
for group in groups:
full_names = tuple(name and (sandbox_path / name)
for name in group.names)
full_groups.append(
SampleGroup(group.enum, full_names, group.project_codes))
return full_groups
def find_sample_groups(run_path, base_calls_path):
# noinspection PyBroadException
try:
fastq_files = base_calls_path.glob("*_R1_*.fastq.gz")
sample_sheet_path = run_path / "SampleSheet.csv"
file_names = [f.name for f in fastq_files]
sample_groups = list(find_groups(file_names, sample_sheet_path))
sample_groups.sort(key=lambda group: get_sample_number(group.names[0]),
reverse=True)
except Exception:
logger.error("Finding sample groups in %s", run_path, exc_info=True)
(run_path / "errorprocessing").write_text(
"Finding sample groups failed.\n")
sample_groups = []
return sample_groups
def find_sample_groups(run_path, base_calls_path):
# noinspection PyBroadException
try:
fastq_files = base_calls_path.glob("*_R1_*.fastq.gz")
sample_sheet_path = run_path / "SampleSheet.csv"
file_names = [f.name for f in fastq_files]
sample_groups = list(find_groups(file_names, sample_sheet_path))
sample_groups.sort(key=lambda group: get_sample_number(group.names[0]),
reverse=True)
except Exception:
logger.error("Finding sample groups in %s", run_path, exc_info=True)
(run_path / "errorprocessing").write_text(
"Finding sample groups failed.\n")
sample_groups = []
return sample_groups
def test_unmatched_midi_file_not_found(tmpdir):
sample_sheet_path = Path(str(tmpdir)) / 'SampleSheet.csv'
sample_sheet_path.write_text(BASIC_HEADER + """\
Sample_ID,Sample_Name,Sample_Plate,Sample_Well,index,index2,Sample_Project,Description,GenomeFolder
CFE_SomeId_10-Jul-2014_N501-N701_Sample1MIDI_MidHCV,Sample1MIDI_MidHCV,,,ACGTACGT,TGCATGCA,,,
CFE_SomeId_10-Jul-2014_N501-N702_Sample2_Proj2,Sample2_Proj2,,,AAAAGGGG,CCCCTTTT,,,
""")
expected_groups = [
SampleGroup('Sample2', ('Sample2-Proj2_S2_L001_R1_001.fastq.gz', None),
('Proj2', None))
]
groups = list(
find_groups(['Sample2-Proj2_S2_L001_R1_001.fastq.gz'],
sample_sheet_path))
assert expected_groups == groups
def test_combine_midi(tmpdir):
sample_sheet_path = Path(str(tmpdir)) / 'SampleSheet.csv'
sample_sheet_path.write_text(BASIC_HEADER + """\
Sample_ID,Sample_Name,Sample_Plate,Sample_Well,index,index2,Sample_Project,Description,GenomeFolder
CFE_SomeId_10-Jul-2014_N501-N701_Sample1_HCV,Sample1_HCV,,,ACGTACGT,TGCATGCA,,,
CFE_SomeId_10-Jul-2014_N501-N702_Sample1MIDI_MidHCV,Sample1MIDI_MidHCV,,,AAAAGGGG,CCCCTTTT,,,
""")
expected_groups = [
SampleGroup('Sample1', ('Sample1-HCV_S1_L001_R1_001.fastq.gz',
'Sample1MIDI-MidHCV_S2_L001_R1_001.fastq.gz'),
('HCV', 'MidHCV'))
]
groups = list(
find_groups([
'Sample1-HCV_S1_L001_R1_001.fastq.gz',
'Sample1MIDI-MidHCV_S2_L001_R1_001.fastq.gz'
], sample_sheet_path))
assert expected_groups == groups
def link_samples(
run_path: str,
output_path: str,
is_denovo: bool,
fastq1s: typing.Sequence[str] = None,
fastq2s: typing.Sequence[str] = None,
project_code: str = None):
""" Load the data from a run folder. """
shutil.rmtree(output_path, ignore_errors=True)
makedirs(output_path)
scratch_path = os.path.join(output_path, 'scratch')
makedirs(scratch_path)
sample_groups = []
run_info_path = os.path.join(run_path, 'RunInfo.xml')
interop_path = os.path.join(run_path, 'InterOp')
if not (os.path.exists(run_info_path) and os.path.exists(interop_path)):
read_sizes = None
else:
read_sizes = parse_read_sizes(run_info_path)
run_info = RunInfo(sample_groups,
reports=['PR_RT', 'IN', 'NS3', 'NS5a', 'NS5b'],
interop_path=interop_path,
scratch_path=scratch_path,
output_path=output_path,
read_sizes=read_sizes,
is_denovo=is_denovo)
sample_sheet_path = os.path.join(run_path, "SampleSheet.csv")
if (fastq1s is not None and len(fastq1s) > 0
or not os.path.exists(sample_sheet_path)):
if fastq1s is not None and len(fastq1s) > 0: # forward files are specified
if fastq2s is None:
raise ValueError("Reverse read files must also be specified.")
elif len(fastq2s) != len(fastq1s):
raise ValueError(
"The same number of forward and reverse read files must be "
"specified."
)
forward_reverse_pairs = zip(fastq1s, fastq2s)
else: # there is no sample sheet
# Sort the FASTQ files alphabetically and run them in pairs.
logger.info(
"No sample sheet found; running on all FASTQ files in folder {}".format(
run_path
)
)
fastq_files = (list(glob(os.path.join(run_path, "*.fastq")))
+ list(glob(os.path.join(run_path, "*.fastq.gz"))))
fastq_files.sort()
forward_reverse_pairs = []
for idx in range(0, len(fastq_files), 2):
forward = fastq_files[idx]
if idx == len(fastq_files) - 1:
# We have an odd number of FASTQ files; ignore this last one.
logger.info(
"File {} appears extraneous; omitting.".format(forward)
)
break
reverse = fastq_files[idx + 1]
logger.info(
"Pairing files {} and {}.".format(forward, reverse)
)
forward_reverse_pairs.append((forward, reverse))
for forward, reverse in forward_reverse_pairs:
sample = Sample(
fastq1=os.path.join(run_path, forward),
fastq2=os.path.join(run_path, reverse),
)
sample.project_code = project_code
sample_groups.append(SampleGroup(sample, midi_sample=None))
else: # a sample sheet is specified
fastq_files = list(glob(os.path.join(run_path,
'Data',
'Intensities',
'BaseCalls',
'*_R1_*')) or
glob(os.path.join(run_path,
'*_R1_*')))
source_folder = fastq_files and os.path.dirname(fastq_files[0])
file_names = [os.path.basename(fastq_file) for fastq_file in fastq_files]
groups = find_groups(file_names, sample_sheet_path)
for group in groups:
main_file, midi_file = group.names
if main_file.startswith('Undetermined'):
continue
main_sample = Sample(fastq1=os.path.join(source_folder, main_file))
main_sample.project_code = project_code
if midi_file is None:
midi_sample = None
else:
midi_sample = Sample(fastq1=os.path.join(source_folder, midi_file))
midi_sample.project_code = project_code
sample_groups.append(SampleGroup(main_sample, midi_sample))
sample_count = sum(1 for _ in run_info.get_all_samples())
for i, sample in enumerate(run_info.get_all_samples(), 1):
sample.rank = '{} of {}'.format(i, sample_count)
sample.bad_cycles_csv = run_info.bad_cycles_csv
sample.scratch_path = os.path.join(scratch_path, sample.name)
return run_info
def genreport_rerun(source, working):
run_path = os.path.dirname(os.path.dirname(source))
run_name = os.path.basename(run_path)
publish_path = os.path.join(working,
'rerun_results',
run_name)
os.makedirs(publish_path)
print('##', run_name)
working_paths = split_files(source, working)
file_names = (os.path.basename(working_path) for working_path in working_paths)
sorted_file_names = sorted(file_names)
sample_sheet_path = os.path.join(source, '../../SampleSheet.csv')
groups = list(find_groups(sorted_file_names,
sample_sheet_path,
included_projects={'HCV'}))
for group in groups:
working_path = os.path.join(working, group.names[0])
if group.names[1] is None:
midi_name = ''
midi_path = working_path
else:
midi_name = group.names[1]
midi_path = os.path.join(
working,
group.names[1])
if not os.path.isdir(midi_path):
midi_name = 'failed MidHCV'
midi_path = working_path
print(working_path, midi_name)
with open(os.path.join(working_path, 'amino.csv')) as amino_csv, \
open(os.path.join(midi_path, 'amino.csv')) as midi_amino_csv, \
open(os.path.join(working_path, 'resistance.csv'), 'w') as resistance_csv, \
open(os.path.join(working_path, 'mutations.csv'), 'w') as mutations_csv, \
open(os.path.join(working_path, 'resistance_fail.csv'), 'w') as resistance_fail_csv:
report_resistance(amino_csv,
midi_amino_csv,
resistance_csv,
mutations_csv,
resistance_fail_csv)
sample_name = os.path.basename(working_path)
with open(os.path.join(working_path, 'resistance.csv')) as resistance_csv, \
open(os.path.join(working_path, 'mutations.csv')) as mutations_csv, \
open(os.path.join(working_path, 'resistance_report.pdf'), 'wb') as resistance_report_csv:
gen_report(resistance_csv,
mutations_csv,
resistance_report_csv,
sample_name=sample_name)
for file_name in ('resistance.csv', 'mutations.csv', 'resistance_fail.csv'):
with open(os.path.join(publish_path, file_name), 'w') as dest:
dest_writer = csv.writer(dest)
for i, group in enumerate(groups):
working_path = os.path.join(working, group.names[0])
sample_name = os.path.basename(working_path)
with open(os.path.join(working_path, file_name), 'r') as source:
source_reader = csv.reader(source)
for j, row in enumerate(source_reader):
if j != 0:
row.insert(0, sample_name)
elif i == 0:
row.insert(0, 'sample')
else:
continue
dest_writer.writerow(row)
