def run_hmmcopy(
bam_file,
corrected_reads_filename,
segments_filename,
parameters_filename,
metrics_filename,
hmmcopy_tar,
cell_id,
hmmparams,
tempdir,
docker_image
):
# generate wig file for hmmcopy
helpers.makedirs(tempdir)
readcount_wig = os.path.join(tempdir, 'readcounter.wig')
corrected_reads = os.path.join(tempdir, 'corrected_reads.csv')
run_correction_hmmcopy(
bam_file,
corrected_reads,
readcount_wig,
hmmparams,
docker_image
)
hmmcopy_tempdir = os.path.join(tempdir, '{}_hmmcopy'.format(cell_id))
helpers.makedirs(hmmcopy_tempdir)
run_hmmcopy_script(
corrected_reads,
hmmcopy_tempdir,
cell_id,
hmmparams,
docker_image
)
hmmcopy_outdir = os.path.join(hmmcopy_tempdir, str(0))
csvutils.rewrite_csv_file(
os.path.join(hmmcopy_outdir, "reads.csv"), corrected_reads_filename,
dtypes=dtypes()['reads']
)
csvutils.rewrite_csv_file(
os.path.join(hmmcopy_outdir, "params.csv"), parameters_filename,
dtypes=dtypes()['params']
)
csvutils.rewrite_csv_file(
os.path.join(hmmcopy_outdir, "segs.csv"), segments_filename,
dtypes=dtypes()['segs']
)
csvutils.rewrite_csv_file(
os.path.join(hmmcopy_outdir, "metrics.csv"), metrics_filename,
dtypes=dtypes()['metrics']
)
helpers.make_tarfile(hmmcopy_tar, hmmcopy_tempdir)
def add_quality(hmmcopy_metrics, alignment_metrics, multipliers, output, training_data, tempdir):
helpers.makedirs(tempdir)
hmmcopy_tables = ['/hmmcopy/metrics/{}'.format(mult) for mult in multipliers]
model = classify.train_classifier(training_data)
feature_names = model.feature_names_
data = classify.load_data(hmmcopy_metrics, alignment_metrics,
hmmcopy_tables, '/alignment/metrics',
feature_names)
for i, (hmmcopy_table, tabledata) in enumerate(data):
intermediate_output = os.path.join(
tempdir, '{}_metrics_with_quality.csv.gz'.format(i)
)
predictions = classify.classify(model, tabledata)
classify.write_to_output(
hmmcopy_metrics,
hmmcopy_table,
intermediate_output,
predictions)
csvutils.prep_csv_files(intermediate_output, output, dtypes=dtypes()['metrics'])
def concatenate_csv(inputs, output, data_type, low_memory=False):
ref_dtypes = None
if data_type:
ref_dtypes = dtypes()[data_type]
if low_memory:
csvutils.concatenate_csv_files_quick_lowmem(inputs,
output,
dtypes=ref_dtypes)
else:
csvutils.concatenate_csv(inputs, output, dtypes=ref_dtypes)
def get_mappability_col(reads, annotated_reads):
reads = csvutils.read_csv_and_yaml(reads, chunksize=100)
alldata = []
for read_data in reads:
read_data['is_low_mappability'] = (read_data['map'] <= 0.9)
alldata.append(read_data)
alldata = pd.concat(alldata)
csvutils.write_dataframe_to_csv_and_yaml(
alldata, annotated_reads, dtypes()['reads'], write_header=True
)
def add_clustering_order(
reads, metrics, output, chromosomes=None, sample_info=None):
"""
adds sample information to metrics in place
"""
order = get_hierarchical_clustering_order(
reads, chromosomes=chromosomes
)
if not sample_info:
sample_info = {}
for cell_id, order in order.items():
if cell_id not in sample_info:
sample_info[cell_id] = {}
sample_info[cell_id]['order'] = order
csvutils.annotate_csv(metrics, sample_info, output, dtypes()['metrics'])
