def count_covariates(picard, dup_align_bam, ref_file, platform,
snp_file):
"""Step 1 of GATK recalibration process -- counting covariates.
"""
out_file = "%s.recal" % os.path.splitext(dup_align_bam)[0]
params = ["-T", "CountCovariates",
"-cov", "ReadGroupCovariate",
"-cov", "QualityScoreCovariate",
"-cov", "CycleCovariate",
"-cov", "DinucCovariate",
"-cov", "TileCovariate",
"-recalFile", out_file,
"-I", dup_align_bam,
"-R", ref_file,
"-l", "INFO",
"-U",
"-OQ",
"--default_platform", platform,
]
if snp_file:
params += ["-B", "dbsnp,VCF,%s" % snp_file]
if not os.path.exists(out_file):
with curdir_tmpdir() as tmp_dir:
picard.run_gatk(params, tmp_dir)
return out_file
def picard_sort(picard, align_bam):
base, ext = os.path.splitext(align_bam)
out_file = "%s-sort%s" % (base, ext)
if not os.path.exists(out_file):
with curdir_tmpdir() as tmp_dir:
opts = [("INPUT", align_bam),
("OUTPUT", out_file),
("TMP_DIR", tmp_dir),
("SORT_ORDER", "coordinate")]
picard.run("SortSam", opts)
return out_file
def picard_fixmate(picard, align_bam):
"""Run Picard's FixMateInformation generating an aligned output file.
"""
base, ext = os.path.splitext(align_bam)
out_file = "%s-sort%s" % (base, ext)
if not os.path.exists(out_file):
with curdir_tmpdir() as tmp_dir:
opts = [("INPUT", align_bam),
("OUTPUT", out_file),
("TMP_DIR", tmp_dir),
("SORT_ORDER", "coordinate")]
picard.run("FixMateInformation", opts)
return out_file
def mark_duplicates(picard, align_bam):
base, ext = os.path.splitext(align_bam)
base = base.replace(".", "-")
dup_bam = "%s-dup%s" % (base, ext)
dup_metrics = "%s-dup.dup_metrics" % base
if not os.path.exists(dup_bam):
with curdir_tmpdir() as tmp_dir:
opts = [("INPUT", align_bam),
("OUTPUT", dup_bam),
("TMP_DIR", tmp_dir),
("METRICS_FILE", dup_metrics)]
picard.run("MarkDuplicates", opts)
return dup_bam
def merge_bam_files(bam_files, work_dir, config):
"""Merge multiple BAM files from a sample into a single BAM for processing.
"""
out_file = os.path.join(work_dir, os.path.basename(bam_files[0]))
if not os.path.exists(out_file):
picard = PicardRunner(config["program"]["picard"])
with utils.curdir_tmpdir() as tmp_dir:
opts = [("OUTPUT", out_file),
("SORT_ORDER", "coordinate"),
("TMP_DIR", tmp_dir)]
for b in bam_files:
opts.append(("INPUT", b))
picard.run("MergeSamFiles", opts)
return out_file
def fastq_to_bam(picard, sample_name, quality_format, read1, read2):
base, ext = os.path.splitext(os.path.basename(read1))
out_file = "%s.bam" % base
if not os.path.exists(out_file):
with curdir_tmpdir() as tmp_dir:
opts = [("FASTQ", read1),
("TMP_DIR", tmp_dir),
("QUALITY_FORMAT", quality_format),
("SAMPLE_NAME", sample_name),
("OUTPUT", out_file)]
if read2:
opts.append(("FASTQ2", read2))
picard.run("FastqToSam", opts)
return out_file
def main(config_file, align_sam, ref_file, fastq_one, fastq_pair=None,
sample_name=""):
with open(config_file) as in_handle:
config = yaml.load(in_handle)
picard = PicardRunner(config["program"]["picard"])
index_ref_file(picard, ref_file)
base_dir = os.path.split(align_sam)[0]
with curdir_tmpdir() as tmp_dir:
out_fastq_bam = picard_fastq_to_bam(picard, fastq_one, fastq_pair,
base_dir, config["algorithm"]["quality_format"],
sample_name, tmp_dir)
out_bam = picard_merge_bam(picard, align_sam, out_fastq_bam,
ref_file, tmp_dir, fastq_pair is not None)
picard_sort(picard, out_bam, tmp_dir)
def indel_realignment(picard, align_bam, ref_file, intervals):
"""Perform realignment of BAM file in specified regions
"""
out_file = "%s-realign.bam" % os.path.splitext(align_bam)[0]
params = ["-T", "IndelRealigner",
"-I", align_bam,
"-R", ref_file,
"-targetIntervals", intervals,
"-o", out_file,
"-l", "INFO",
]
if not (os.path.exists(out_file) and os.path.getsize(out_file) > 0):
with curdir_tmpdir() as tmp_dir:
picard.run_gatk(params, tmp_dir)
return out_file
def gatk_recalibrate(picard, dup_align_bam, ref_file, recal_file, platform):
"""Step 2 of GATK recalibration -- use covariates to re-write output file.
"""
out_file = "%s-gatkrecal.bam" % os.path.splitext(dup_align_bam)[0]
params = ["-T", "TableRecalibration",
"-recalFile", recal_file,
"-R", ref_file,
"-I", dup_align_bam,
"-outputBam", out_file,
"-l", "INFO",
"-U",
"-OQ",
"--default_platform", platform,
]
if not os.path.exists(out_file):
with curdir_tmpdir() as tmp_dir:
picard.run_gatk(params, tmp_dir)
return out_file
def main(config_file, align_sam, ref_file, fastq_one, fastq_pair=None,
sample_name="", rg_name="", pu_name=""):
with open(config_file) as in_handle:
config = yaml.load(in_handle)
picard = PicardRunner(config["program"]["picard"])
platform = config["algorithm"]["platform"]
if platform.lower() == "illumina":
qual_format = "Illumina"
else:
raise ValueError("Need to specify quality format for %s" % platform)
index_ref_file(picard, ref_file)
base_dir = os.path.split(align_sam)[0]
with curdir_tmpdir() as tmp_dir:
out_fastq_bam = picard_fastq_to_bam(picard, fastq_one, fastq_pair,
base_dir, platform, qual_format, sample_name, rg_name, pu_name,
tmp_dir)
out_bam = picard_merge_bam(picard, align_sam, out_fastq_bam,
ref_file, tmp_dir, fastq_pair is not None)
picard_sort(picard, out_bam, tmp_dir)
def picard_run_maq(picard, maq_cmd, input_bam, ref_file, barcode, lane,
out_base, stringency, is_paired=False, limit=None, ext=""):
out_dir = "%s-maq%s" % (out_base, ext)
if not os.path.exists(out_dir):
os.makedirs(out_dir)
bam_out_file = "%s.bam" % (out_dir)
with curdir_tmpdir() as tmp_dir:
std_opts = [("INPUT", input_bam),
("ANALYSIS_DIR", out_dir),
("FLOWCELL_BARCODE", barcode),
("LANE", lane),
("REFERENCE_SEQUENCE", ref_file),
("TMP_DIR", tmp_dir),
("PAIRED_RUN", ("true" if is_paired else "false"))]
# Convert fastq to Maq ready files
if len(glob.glob(
os.path.join(out_dir, "%s.%s*bfq" % (barcode, lane)))) == 0:
opts = std_opts + [
("PREPARE", "true"),
]
if limit:
opts.append(("READS_TO_ALIGN", int(limit)))
picard.run("RunMaq", opts)
# actually run Maq. Use python as Picard is failing with same parameters
if len(glob.glob(
os.path.join(out_dir, "%s.%s*out*.map" % (barcode, lane)))) == 0:
#opts = std_opts + [
# ("STRINGENCY", stringency),
# ("ALIGN", "true"),
# ]
#picard.run("RunMaq", opts)
run_maq(maq_cmd, stringency, out_dir, ref_file, barcode, lane)
# Convert the output file to BAM aligned
if not os.path.exists(bam_out_file):
opts = std_opts + [
("OUTPUT", bam_out_file),
("BAM_OUTPUT", "true")
]
index_file = index_ref_file(picard, ref_file)
picard.run("RunMaq", opts)
#convert_map_to_bam(picard, out_dir, bam_out_file, ref_file,
# barcode, lane)
return bam_out_file
