def preprocess_reads(args):
"""
Preprocesses fastq files by removing UMIs from reads and appending
them to the read names.
"""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
#config.read('../config/demo_config.ini')
config_validation(conf_paths=dict(
config.items('PATHS'))) ##Check whether PATHS in config file exist
else:
config = None
prepfile = handle_arg(args.prepfile,
config['PATHS']['prep_file'] if config else None,
'ERR: No prepfile provided in args or config.')
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'ERR: No output path provided in args or config.')
arg_exists(sys.argv) ##Check whether args directories/files exist
reheader_fastqs(r1_file=args.read1,
r2_file=args.read2,
r3_file=args.read3,
output_path=output_path,
prepname=args.prepname,
prepfile=prepfile)
def collapse(args):
"""Base collapses from given BAM and umi family file."""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
config_validation(conf_paths=dict(
config.items('PATHS'))) ##Check whether PATHS in config file exist
else:
config = None
region = args.region
if any(item not in region for item in ["chr", ":", "-"]):
raise ValueError(
'ERR: Incorrect region string (should look like chr1:1200000-1250000).'
)
sys.exit(1)
contig = region.split(":")[0]
region_start = int(region.split(":")[1].split("-")[0])
region_end = int(region.split(":")[1].split("-")[1])
bam_file = handle_arg(args.bam_file,
config['PATHS']['bam_file'] if config else None,
'ERR: No BAM file provided in args or config.')
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'ERR: No output path provided in args or config.')
arg_exists(sys.argv) ##Check whether args directories/files exist
if args.umi_file:
umi_file = args.umi_file
elif config:
umi_file = config['PATHS']['umi_file'] if 'umi_file' in config[
'PATHS'] else None
if umi_file:
try:
umi_table = pickle.load(open(umi_file, "rb"))
except IOError:
print("ERR: Unable to load .umis file.", file=sys.stderr)
sys.exit(1)
else:
umi_table = None
print(timestamp() + "Generating consensus...")
generate_consensus_output(contig=contig,
region_start=region_start,
region_end=region_end,
bam_file=bam_file,
umi_table=umi_table,
output_path=output_path,
config=config)
print(timestamp() +
"Consensus generated. Consensus file written to {}.".format(
output_path))
def preprocess_reads(args):
"""
Preprocesses fastq files by removing UMIs from reads and appending
them to the read names.
"""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
else:
config = None
prepfile = handle_arg(args.prepfile,
config['PATHS']['prep_file'] if config else None,
'ERR: No prepfile provided in args or config.')
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'ERR: No output path provided in args or config.')
reheader_fastqs(r1_file=args.read1,
r2_file=args.read2,
r3_file=args.read3,
output_path=output_path,
prepname=args.prepname,
prepfile=prepfile)
def group_umis(args):
"""Groups and error-corrects UMIs into families."""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
else:
config = None
region = args.region
if any(item not in region for item in [":", "-"]):
raise ValueError(
'ERR: Incorrect region string (should look like chr1:1200000-1250000).'
)
sys.exit(1)
contig = region.split(":")[0]
region_start = int(region.split(":")[1].split("-")[0])
region_end = int(region.split(":")[1].split("-")[1])
bam_file = handle_arg(args.bam_file,
config['PATHS']['bam_file'] if config else None,
'ERR: No BAM file provided in args or config.')
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'ERR: No output path provided in args or config.')
print(timestamp() + "Grouping UMIs...")
## Generate an error-corrected list of UMI families
umi_families = get_umi_families(contig=contig,
region_start=region_start,
region_end=region_end,
bam_file=bam_file,
config=config)
umi_file = "{}/{}.umis".format(output_path, region)
pickle.dump(umi_families, open(umi_file, "wb"))
print(timestamp() +
"UMI grouping complete. Output written to {}.".format(output_path))
def call_variants(args):
"""Generates VCF files from given cons file."""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
config_validation(conf_paths=dict(
config.items('PATHS'))) ##Check whether PATHS in config file exist
else:
config = None
cons_file = args.cons_file
f_sizes = args.f_sizes.split(',')
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'No output path provided in args or config.')
region = args.region
arg_exists(sys.argv) ##Check whether args directories/files exist
cons_is_merged = check_consfile(cons_file)
if cons_is_merged:
region_start = region.split("_")[0]
region_end = region.split("_")[1]
else:
if any(x not in region for x in ["chr", ":", "-"]):
raise ValueError(
'Incorrect region string (should look like chr1:1200000-1250000).'
)
sys.exit(1)
contig = region.split(":")[0]
region_start = int(region.split(":")[1].split("-")[0])
region_end = int(region.split(":")[1].split("-")[1])
print(timestamp() + "Generating VCFs...")
get_vcf_output(cons_file=cons_file,
region_start=region_start,
region_end=region_end,
output_path=output_path,
config=config,
run_id=run_id)
print(timestamp() +
"VCFs generated. VCF files written to {}.".format(output_path))
def call_variants(args):
"""Generates VCF files from given cons file."""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
else:
config = None
cons_file = args.cons_file
f_sizes = args.f_sizes.split(',')
region = args.region
if any(x not in region for x in [":", "-"]):
raise ValueError(
'Incorrect region string (should look like chr1:1200000-1250000).')
sys.exit(1)
contig = region.split(":")[0]
region_start = int(region.split(":")[1].split("-")[0])
region_end = int(region.split(":")[1].split("-")[1])
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'No output path provided in args or config.')
print(timestamp() + "Generating VCFs...")
generate_vcf_output(cons_file=cons_file,
f_sizes=f_sizes,
contig=contig,
region_start=region_start,
region_end=region_end,
output_path=output_path,
config=config)
print(timestamp() +
"VCFs generated. VCF files written to {}.".format(output_path))
parser.add_argument('-o', '--output_path', help='Path to write output files to.')
parser.add_argument('-c', '--config', help='Path to your config file.')
parser.add_argument('-t', '--tally', help='Path to your tally (output of UMI_count.py).')
args = parser.parse_args()
if args.config:
config = configparser.ConfigParser()
config.read(config_file)
else:
config = None
region = args.region
if any(x not in region for x in ["chr", ":", "-"]):
raise ValueError('Incorrect region string (should look like chr1:1200000-1250000).')
sys.exit(1)
contig = region.split(":")[0]
region_start = int(region.split(":")[1].split("-")[0])
region_end = int(region.split(":")[1].split("-")[1])
bam_file = handle_arg(args.bam_file, config['PATHS']['bam_file'] if config else None,
'No BAM file provided in args or config.')
output_path = handle_arg(args.output_path, config['PATHS']['output_path'] if config else None,
'No output path provided in args or config.')
tally_file = handle_arg(args.tally, output_path + '/' + region + '.tally' if config else None,
'No tally file provided.')
## Output
generate_consensus_output(contig, region_start, region_end, bam_file, tally_file, output_path, config)
def group_umis(args):
"""Groups and error-corrects UMIs into families."""
if args.config:
config = configparser.ConfigParser()
config.read(args.config)
config_validation(conf_paths=dict(
config.items('PATHS'))) ##Check whether PATHS in config file exist
else:
config = None
region = args.region
if any(item not in region for item in ["chr", ":", "-"]):
raise ValueError(
'ERR: Incorrect region string (should look like chr1:1200000-1250000).'
)
sys.exit(1)
contig = region.split(":")[0]
region_start = int(region.split(":")[1].split("-")[0])
region_end = int(region.split(":")[1].split("-")[1])
bam_file = handle_arg(args.bam_file,
config['PATHS']['bam_file'] if config else None,
'ERR: No BAM file provided in args or config.')
output_path = handle_arg(
args.output_path, config['PATHS']['output_path'] if config else None,
'ERR: No output path provided in args or config.')
arg_exists(sys.argv) ##Check whether args directories/files exist
print(timestamp() + "Grouping UMIs...")
## Generate an error-corrected list of UMI families
umi_families, umi_groups = get_umi_families(contig=contig,
region_start=region_start,
region_end=region_end,
bam_file=bam_file,
config=config)
total_parent_umi_count, total_child_umi_count, num_of_children, freq_of_parent_umis = umi_datafile(
umi_groups)
filename = "{}/datafile_{}.csv".format(output_path, region)
headers = [
'CHR', 'START', 'END', 'PTU', 'CTU', 'CHILD_NUMS', 'FREQ_PARENTS'
]
csv.register_dialect('myDialect', delimiter='\t', quoting=csv.QUOTE_NONE)
csvrow = {
'CHR': contig,
'START': str(region_start),
'END': str(region_end),
'PTU': str(total_parent_umi_count),
'CTU': str(total_child_umi_count),
'CHILD_NUMS': num_of_children,
'FREQ_PARENTS': freq_of_parent_umis
}
info = [
contig, region_start, region_end, total_parent_umi_count,
total_child_umi_count, num_of_children, freq_of_parent_umis
]
file = open(filename, "w")
writer = csv.DictWriter(file, dialect='myDialect', fieldnames=headers)
writer.writeheader()
writer.writerow(csvrow)
umi_file = "{}/{}.umis".format(output_path, region)
pickle.dump(umi_families, open(umi_file, "wb"))
print(timestamp() +
"UMI grouping complete. Output written to {}.".format(output_path))