def main(config_file):
with open(config_file) as in_handle:
config = yaml.load(in_handle)
ref_index = novoalign.refindex(config["ref"], kmer_size=13, step_size=1)
create_dirs(config)
for cur in config["input"]:
in_fastq = cur["fastq"]
if cur.get("old_style_barcodes", False):
in_fastq = convert_illumina_oldstyle(in_fastq)
bc_files = demultiplex(in_fastq, cur["barcodes"],
config["dir"]["tmp"], config)
with cpmap(config["algorithm"]["cores"]) as cur_map:
for _ in cur_map(process_fastq, ((bc_file, ref_index, cur, config, config_file)
for bc_file in bc_files)):
pass
def fastq_to_process(config):
"""Retrieve fastq files to process, handling demultiplexing.
"""
for cur in config.get("experiments", config.get("input", [])):
ref_index = _index_ref_genome(config, cur)
in_fastq = cur.get("fastq", None)
if cur.get("old_style_barcodes", False):
in_fastq = convert_illumina_oldstyle(in_fastq)
if cur.get("barcodes", None):
bc_files = demultiplex(in_fastq, [(b["name"], b["seq"]) for b in cur["barcodes"]],
config["dir"]["tmp"], config)
fastqs = _assign_bc_files(cur, bc_files)
else:
cur["program"] = config["program"]
if not cur.has_key("algorithm"):
cur["algorithm"] = config["algorithm"]
fastqs = [cur]
for fq in fastqs:
yield fq, ref_index