def process_fastq(curinfo, ref_index, config, config_file):
do_realignment = config["algorithm"].get("realignment", "")
do_kmercorrect = config["algorithm"].get("kmer_correct", "")
trim_three = config["algorithm"].get("trim_three", "")
picard = broad.runner_from_config(config)
in_file = _prepare_fastq(curinfo, config)
if trim_three:
in_file = trim_fastq(in_file, three=int(trim_three))
if do_kmercorrect:
in_file = remove_ns(in_file)
in_file = kmer_filter(in_file, do_kmercorrect, config)
unique_file, count_file = uniquify_reads(in_file, config)
align_sam = novoalign.align(unique_file, None, ref_index,
os.path.splitext(os.path.basename(in_file))[0],
config["dir"]["align"],
curinfo)
name = curinfo.get("description", curinfo.get("name", ""))
ref = curinfo.get("ref", config.get("ref", None))
align_bam = sam_to_sort_bam(align_sam, ref, unique_file, None,
name, name, name, config)
if do_realignment == "gatk":
align_bam = gatk_realigner(align_bam, ref, config, deep_coverage=True)
picard.run_fn("picard_index", align_bam)
# XXX Finish remainder of processing
summarize_at_each_pos(align_bam, in_file, count_file, name, config)
return
if config["algorithm"].get("range_params", None):
call_analyze_multiple(align_bam, bc, in_file, config)
else:
call_bases_and_analyze(align_bam, bc, in_file, config)
def process_fastq(curinfo, ref_index, config, config_file):
do_realignment = config["algorithm"].get("realignment", "")
do_kmercorrect = config["algorithm"].get("kmer_correct", "")
trim_three = config["algorithm"].get("trim_three", "")
picard = broad.runner_from_config(config)
in_file = _prepare_fastq(curinfo, config)
if trim_three:
in_file = trim_fastq(in_file, three=int(trim_three))
if do_kmercorrect:
in_file = remove_ns(in_file)
in_file = kmer_filter(in_file, do_kmercorrect, config)
unique_file, count_file = uniquify_reads(in_file, config)
align_sam = novoalign.align(unique_file, None, ref_index,
os.path.splitext(os.path.basename(in_file))[0],
config["dir"]["align"],
curinfo)
name = curinfo.get("description", curinfo.get("name", ""))
ref = curinfo.get("ref", config.get("ref", None))
names = {"rg": name, "sample": name, "pu": name, "pl": name}
align_bam = sam_to_sort_bam(align_sam, ref, unique_file, None,
names, config)
if do_realignment == "gatk":
align_bam = gatk_realigner(align_bam, ref, config, deep_coverage=True)
picard.run_fn("picard_index", align_bam)
# XXX Finish remainder of processing
summarize_at_each_pos(align_bam, in_file, count_file, name, config)
def align_and_sort(in_file, exp, config, align_dir):
out_sam = os.path.join(align_dir, "%s.sam" % (os.path.splitext(
os.path.basename(in_file))[0]))
out_bam = "%s.bam" % os.path.splitext(out_sam)[0]
sort_bam = "%s-sort.bam" % os.path.splitext(out_sam)[0]
if not os.path.exists(sort_bam):
if not os.path.exists(out_sam):
cl = [config["programs"]["bowtie"],
"-S", "-f", "--all", "--best", "--strata",
"-v", str(config["analysis"]["align_errors"]),
config["analysis"]["bowtie_genome"],
in_file, out_sam]
print cl
subprocess.check_call(cl)
sort_bam = sam_to_sort_bam(out_sam, config["analysis"]["seq_genome"],
in_file, None, exp["name"], "", exp["name"],
config)
for to_remove in [out_sam, out_bam]:
if os.path.exists(to_remove):
os.remove(to_remove)
return sort_bam
def process_fastq(bc, ref_index, cur_config, config, config_file):
do_realignment = config["algorithm"].get("realignment", "")
do_kmercorrect = config["algorithm"].get("kmer_correct", "")
trim_three = config["algorithm"].get("trim_three", "")
picard = broad.runner_from_config(config)
in_file = bc["file"]
if trim_three:
in_file = trim_fastq(in_file, three=int(trim_three))
if do_kmercorrect:
in_file = remove_ns(in_file)
in_file = kmer_filter(in_file, do_kmercorrect, config)
unique_file = uniquify_bioplayground(in_file, config)
align_sam = novoalign.align(config["dir"]["align"], ref_index, unique_file,
qual_format=cur_config.get("format", None))
align_bam = sam_to_sort_bam(align_sam, config["ref"], unique_file, None,
"", bc["name"], config)
if do_realignment == "gatk":
align_bam = gatk_realigner(align_bam, config["ref"], config, deep_coverage=True)
picard.run_fn("picard_index", align_bam)
if config["algorithm"].get("range_params", None):
call_analyze_multiple(align_bam, bc, in_file, config)
else:
call_bases_and_analyze(align_bam, bc, in_file, config)
