accession = lib['bioSple']
print(accession)
total_reads = 0
for batch_lib in batch_obj['batch']['readsDist']['DNA']:
if batch_lib['bioSple'] == accession:
total_reads = batch_lib['postQualityReads']
print(total_reads)
if total_reads == 0:
continue
summary_paths = lib['diagnosticOutput']
composition_path = glob.glob(
os.path.join(results_dir, 'tax',
seq_sple + '*sample_composition.out'))
if composition_path:
sample_comp = bin_reads(composition_path[0],
ncbi_class=ncbi,
ctrl_taxids=[10760],
quantification='relative')
# for key in sample_comp:
# sample_comp[key] = sample_comp[key] * 10e6 / total_reads
else:
sample_comp = {
"Human": None,
"Bacteria": None,
"Virus": None,
"Parasite": None,
"Fungus": None,
"Unclassified": None
}
viral_summary = [
path for path in lib['diagnosticOutput']
if 'dna.viral.dxsm.out.summary.gz' in path
def get_lib_dict_ls(results_dir=None, fqo=None):
fqo_hi_ic = pd.read_csv(fqo)
fqo_hi_ic['Run Date (YYYY-MM-DD)'] = pd.to_datetime(
fqo_hi_ic['Run Date (YYYY-MM-DD)'])
results_dir = results_dir
batch_paths = glob.glob(os.path.join(results_dir, 'batch', '*'))
lib_dict_ls = []
for batch_path in batch_paths:
with open(batch_path) as batch_file:
batch = json.load(batch_file)
batch_id = batch['batch']['libBatchId']
for lib in batch['libraries']:
accession = lib['bioSple']
seq_sple = lib['seqSple']
lib_type = lib['libType']
if not any(fqo_hi_ic['Seq Sple'] == seq_sple.lower()):
continue
run_date = fqo_hi_ic.loc[fqo_hi_ic['Seq Sple'] == seq_sple.lower(),
'Run Date (YYYY-MM-DD)'].values[0]
sample_name = lib['spleName']
total_reads = lib['qualityFilterInfo']['readsOut']
tax_paths = lib['diagnosticOutput']
vir_paths = [
path for path in tax_paths
if 'dna.viral.dxsm.out.summary.gz' in path
]
t7_read_cnt = 0
pr772_read_cnt = 0
if vir_paths:
vir_path = vir_paths[0]
with gzip.open(os.path.join(results_dir,
vir_path)) as vir_file:
for line in vir_file:
obj = json.loads(line.strip())
if obj['reporting_id'] == '26706_10760':
t7_read_cnt = obj['read_count']
elif obj['reporting_id'] == '26648_261665':
pr772_read_cnt = obj['read_count']
t7_norm = 10e6 * t7_read_cnt / total_reads
pr772_norm = 10e6 * pr772_read_cnt / total_reads
# composition file data
composition_path_ls = glob.glob(
os.path.join(results_dir, 'tax',
seq_sple + '*dna.sample_composition.out'))
if not composition_path_ls:
composition_data = {
'Total Read Count': 1e-6,
'Taxid Read Count': [(10760, 0), (261665, 0)]
}
org_composition = {
"Human": 0,
"Bacteria": 0,
"Virus": 0,
"Parasite": 0,
"Fungus": 0,
"Unclassified": 0
}
composition_file = None
else:
composition_path = composition_path_ls[0]
composition_file = os.path.basename(composition_path)
composition_data = parse_composition_file(
composition_path, [10760, 261665])
org_composition = bin_reads(composition_path,
ncbi_class=ncbi,
quantification='relative',
ctrl_taxids=[10760, 261665])
# for key in org_composition:
# org_composition[key] = 10e6 * org_composition[key] / composition_data['Total Read Count']
if not composition_data['Taxid Read Count']:
composition_data['Taxid Read Count'] = [(10760, 0),
(261665, 0)]
elif len(composition_data['Taxid Read Count']) == 1:
if composition_data['Taxid Read Count'][0][0] == 10760:
composition_data['Taxid Read Count'].append((261665, 0))
elif composition_data['Taxid Read Count'][0][0] == 261665:
composition_data['Taxid Read Count'].append((10760, 0))
t7_raw_comp = [
item[1] for item in composition_data['Taxid Read Count']
if item[0] == 10760
][0]
pr772_raw_comp = [
item[1] for item in composition_data['Taxid Read Count']
if item[0] == 261665
][0]
t7_norm_comp = 10e6 * t7_raw_comp / composition_data[
'Total Read Count']
pr772_norm_comp = 10e6 * pr772_raw_comp / composition_data[
'Total Read Count']
lib_dict = {
'Accession':
accession,
'Seq Sple':
seq_sple,
'Composition File':
composition_file,
'Batch ID':
batch_id,
'Run Date':
run_date,
'Sample Name':
sample_name,
'Library Type':
lib_type,
'Total Reads':
total_reads,
'T7 Raw Reads':
t7_read_cnt,
'PR772 Raw Reads':
pr772_read_cnt,
'T7 Normalized Reads':
t7_norm,
'PR772 Normalized Reads':
pr772_norm,
'T7 + PR772 NR':
t7_norm + pr772_norm,
'Log10 T7 + PR772 NR':
np.log10(t7_norm + pr772_norm),
'Total Reads (Composition File)':
composition_data['Total Read Count'],
'T7 Raw Reads (Composition File)':
t7_raw_comp,
'T4 Raw Reads (Composition File)':
pr772_raw_comp,
'T7 Normalized (Composition File)':
t7_norm_comp,
'T4 Normalized (Composition File)':
pr772_norm_comp,
'T7 + PR772 NR (Composition File)':
t7_norm_comp + pr772_norm_comp,
'Log10 T7 + PR772 NR (Composition File)':
np.log10(t7_norm_comp + pr772_norm_comp),
'Summary - Composition (log10)':
np.log10(t7_norm + pr772_norm) -
np.log10(t7_norm_comp + pr772_norm_comp),
"Human":
org_composition['Human'],
"Bacteria":
org_composition['Bacteria'],
"Virus":
org_composition['Virus'],
"Parasite":
org_composition['Parasite'],
"Fungus":
org_composition['Fungus'],
"Unclassified":
org_composition['Unclassified']
}
lib_dict_ls.append(lib_dict)
return lib_dict_ls
batch_file_paths = glob.glob(os.path.join(results_dir, 'batch', '*'))
results = {}
for batch_file in batch_file_paths:
with open(batch_file) as input:
batch_obj = json.load(input)
for lib in batch_obj['libraries']:
seq_sple = lib['seqSple']
accession = lib['bioSple']
for batch_lib in batch_obj['batch']['readsDist']['DNA']:
if batch_lib['bioSple'] == accession:
total_reads = batch_lib['postQualityReads']
summary_paths = lib['diagnosticOutput']
composition_path = glob.glob(os.path.join(results_dir, 'tax', seq_sple + '*sample_composition.out'))
if composition_path:
sample_comp = bin_reads(composition_path[0], ncbi_class=ncbi, quantification='absolute', ctrl_taxids=[10760])
for key in sample_comp:
sample_comp[key] = sample_comp[key] * 10e6 / total_reads
else:
sample_comp = {
"Human": None,
"Bacteria": None,
"Virus": None,
"Parasite": None,
"Fungus": None,
"Unclassified": None
}
viral_summary = [path for path in lib['diagnosticOutput'] if 'dna.viral.dxsm.out.summary.gz' in path]
if viral_summary:
viral_summary_path = os.path.join(results_dir, viral_summary[0])
with gzip.open(viral_summary_path, 'rt') as viral_summary_file:
os.path.join('low_ic_compositions', low_ic_comp), [10760, 261665])
low_ic_t7_raw_comp = [
item[1] for item in low_ic_comp_data['Taxid Read Count']
if item[0] == 10760
][0]
low_ic_pr772_raw_comp = [
item[1] for item in low_ic_comp_data['Taxid Read Count']
if item[0] == 261665
][0]
low_ic_t7_norm_comp = 10e6 * low_ic_t7_raw_comp / low_ic_comp_data[
'Total Read Count']
low_ic_pr772_norm_comp = 10e6 * low_ic_pr772_raw_comp / low_ic_comp_data[
'Total Read Count']
org_composition = bin_reads(os.path.join('low_ic_compositions',
low_ic_comp),
ncbi_class=ncbi,
quantification='relative',
ctrl_taxids=[10760, 261665])
# for key in org_composition:
# org_composition[key] = 10e6 * org_composition[key] / low_ic_comp_data['Total Read Count']
low_ic_comp_dict = {
'Accession':
accession,
'Seq Sple':
seq_sple,
'Composition File':
low_ic_comp,
'Batch ID':
batch_id,
'Run Date':
run_date,