def align_transcriptome(fastq_file, pair_file, ref_file, data):
"""
bwa mem with settings for aligning to the transcriptome for eXpress/RSEM/etc
"""
work_bam = dd.get_work_bam(data)
base, ext = os.path.splitext(work_bam)
out_file = base + ".transcriptome" + ext
if utils.file_exists(out_file):
data = dd.set_transcriptome_bam(data, out_file)
return data
# bwa mem needs phred+33 quality, so convert if it is Illumina
if dd.get_quality_format(data).lower() == "illumina":
logger.info("bwa mem does not support the phred+64 quality format, "
"converting %s and %s to phred+33.")
fastq_file = fastq.groom(fastq_file, data, in_qual="fastq-illumina")
if pair_file:
pair_file = fastq.groom(pair_file, data, in_qual="fastq-illumina")
bwa = config_utils.get_program("bwa", data["config"])
gtf_file = dd.get_gtf_file(data)
gtf_fasta = index_transcriptome(gtf_file, ref_file, data)
args = " ".join(_bwa_args_from_config(data["config"]))
num_cores = data["config"]["algorithm"].get("num_cores", 1)
samtools = config_utils.get_program("samtools", data["config"])
cmd = ("{bwa} mem {args} -a -t {num_cores} {gtf_fasta} {fastq_file} "
"{pair_file} | {samtools} view -bhS - > {tx_out_file}")
with file_transaction(data, out_file) as tx_out_file:
message = "Aligning %s and %s to the transcriptome." % (fastq_file,
pair_file)
do.run(cmd.format(**locals()), message)
data = dd.set_transcriptome_bam(data, out_file)
return data
def align_transcriptome(fastq_file, pair_file, ref_file, data):
"""
bwa mem with settings for aligning to the transcriptome for eXpress/RSEM/etc
"""
work_bam = dd.get_work_bam(data)
base, ext = os.path.splitext(work_bam)
out_file = base + ".transcriptome" + ext
if utils.file_exists(out_file):
data = dd.set_transcriptome_bam(data, out_file)
return data
# bwa mem needs phred+33 quality, so convert if it is Illumina
if dd.get_quality_format(data).lower() == "illumina":
logger.info("bwa mem does not support the phred+64 quality format, " "converting %s and %s to phred+33.")
fastq_file = fastq.groom(fastq_file, in_qual="fastq-illumina", data=data)
if pair_file:
pair_file = fastq.groom(pair_file, in_qual="fastq-illumina", data=data)
bwa = config_utils.get_program("bwa", data["config"])
gtf_file = dd.get_gtf_file(data)
gtf_fasta = index_transcriptome(gtf_file, ref_file, data)
args = " ".join(_bwa_args_from_config(data["config"]))
num_cores = data["config"]["algorithm"].get("num_cores", 1)
cmd = (
"{bwa} mem {args} -a -t {num_cores} {gtf_fasta} {fastq_file} "
"{pair_file} | samtools view -bhS - > {tx_out_file}"
)
with file_transaction(out_file) as tx_out_file:
message = "Aligning %s and %s to the transcriptome." % (fastq_file, pair_file)
do.run(cmd.format(**locals()), message)
data = dd.set_transcriptome_bam(data, out_file)
return data
def align_transcriptome(fastq_file, pair_file, ref_file, data):
"""
bowtie2 with settings for aligning to the transcriptome for eXpress/RSEM/etc
"""
work_bam = dd.get_work_bam(data)
base, ext = os.path.splitext(work_bam)
out_file = base + ".transcriptome" + ext
if utils.file_exists(out_file):
data = dd.set_transcriptome_bam(data, out_file)
return data
bowtie2 = config_utils.get_program("bowtie2", data["config"])
gtf_file = dd.get_gtf_file(data)
gtf_index = index_transcriptome(gtf_file, ref_file, data)
num_cores = data["config"]["algorithm"].get("num_cores", 1)
fastq_cmd = "-1 %s" % fastq_file if pair_file else "-U %s" % fastq_file
pair_cmd = "-2 %s " % pair_file if pair_file else ""
cmd = (
"{bowtie2} -p {num_cores} -a -X 600 --rdg 6,5 --rfg 6,5 --score-min L,-.6,-.4 --no-discordant --no-mixed -x {gtf_index} {fastq_cmd} {pair_cmd} "
)
with file_transaction(data, out_file) as tx_out_file:
message = "Aligning %s and %s to the transcriptome." % (fastq_file,
pair_file)
cmd += "| " + postalign.sam_to_sortbam_cl(
data, tx_out_file, name_sort=True)
do.run(cmd.format(**locals()), message)
data = dd.set_transcriptome_bam(data, out_file)
return data
def _update_data(align_file, out_dir, names, data):
data = dd.set_work_bam(data, align_file)
data = dd.set_align_bam(data, align_file)
if dd.get_transcriptome_align(data) and not is_transcriptome_broken():
transcriptome_file = _move_transcriptome_file(out_dir, names)
data = dd.set_transcriptome_bam(data, transcriptome_file)
return data
def _update_data(align_file, out_dir, names, data):
data = dd.set_work_bam(data, align_file)
data = dd.set_align_bam(data, align_file)
if dd.get_transcriptome_align(data) and not is_transcriptome_broken(data):
transcriptome_file = _move_transcriptome_file(out_dir, names)
data = dd.set_transcriptome_bam(data, transcriptome_file)
return data
def _update_data(align_file, out_dir, names, data):
data = dd.set_work_bam(data, align_file)
data = dd.set_align_bam(data, align_file)
transcriptome_file = _move_transcriptome_file(out_dir, names)
data = dd.set_transcriptome_bam(data, transcriptome_file)
sjfile = get_splicejunction_file(out_dir, data)
sjbed = junction2bed(sjfile)
data = dd.set_junction_bed(data, sjbed)
return data
def _update_data(align_file, out_dir, names, data):
data = dd.set_work_bam(data, align_file)
data = dd.set_align_bam(data, align_file)
transcriptome_file = _move_transcriptome_file(out_dir, names)
data = dd.set_transcriptome_bam(data, transcriptome_file)
sjfile = get_splicejunction_file(out_dir, data)
if sjfile:
sjbed = junction2bed(sjfile)
data = dd.set_junction_bed(data, sjbed)
return data
def align_transcriptome(fastq_file, pair_file, ref_file, data):
"""
bowtie2 with settings for aligning to the transcriptome for eXpress/RSEM/etc
"""
work_bam = dd.get_work_bam(data)
base, ext = os.path.splitext(work_bam)
out_file = base + ".transcriptome" + ext
if file_exists(out_file):
data = dd.set_transcriptome_bam(data, out_file)
return data
bowtie2 = config_utils.get_program("bowtie2", data["config"])
gtf_file = dd.get_gtf_file(data)
gtf_index = index_transcriptome(gtf_file, ref_file, data)
num_cores = data["config"]["algorithm"].get("num_cores", 1)
fastq_cmd = "-1 %s" % fastq_file if pair_file else "-U %s" % fastq_file
pair_cmd = "-2 %s " % pair_file if pair_file else ""
cmd = ("{bowtie2} -p {num_cores} -a -X 600 --rdg 6,5 --rfg 6,5 --score-min L,-.6,-.4 --no-discordant --no-mixed -x {gtf_index} {fastq_cmd} {pair_cmd} | samtools view -hbS - > {tx_out_file}")
with file_transaction(out_file) as tx_out_file:
message = "Aligning %s and %s to the transcriptome." % (fastq_file, pair_file)
do.run(cmd.format(**locals()), message)
data = dd.set_transcriptome_bam(data, out_file)
return data
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data["oncofuse_file"] = oncofuse.run(data)
if dd.get_dexseq_gff(data, None):
data = dd.set_dexseq_counts(data, dexseq.bcbio_run(data))
# if RSEM was run, stick the transcriptome BAM file into the datadict
if dd.get_aligner(data).lower() == "star" and dd.get_rsem(data):
base, ext = os.path.splitext(dd.get_work_bam(data))
data = dd.set_transcriptome_bam(data, base + ".transcriptome" + ext)
return [[data]]
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data["oncofuse_file"] = oncofuse.run(data)
if dd.get_dexseq_gff(data, None):
data = dd.set_dexseq_counts(data, dexseq.bcbio_run(data))
# if RSEM was run, stick the transcriptome BAM file into the datadict
if dd.get_aligner(data).lower() == "star" and dd.get_rsem(data):
base, ext = os.path.splitext(dd.get_work_bam(data))
data = dd.set_transcriptome_bam(data, base + ".transcriptome" + ext)
return [[data]]
def run_rapmap_align(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
rapmap_dir = os.path.join(work_dir, "rapmap", samplename)
gtf_file = dd.get_gtf_file(data)
fasta_file = dd.get_ref_file(data)
out_file = rapmap_align(fq1, fq2, rapmap_dir, gtf_file, fasta_file,
"quasi", data)
data = dd.set_transcriptome_bam(data, out_file)
return [[data]]
def run_rapmap_align(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
rapmap_dir = os.path.join(work_dir, "rapmap", samplename)
gtf_file = dd.get_gtf_file(data)
fasta_file = dd.get_ref_file(data)
out_file = rapmap_align(fq1, fq2, rapmap_dir, gtf_file, fasta_file,
"quasi", data)
data = dd.set_transcriptome_bam(data, out_file)
return [[data]]
def run_rapmap_pseudoalign(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
rapmap_dir = os.path.join(work_dir, "rapmap", samplename)
gtf_file = dd.get_gtf_file(data)
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
out_file = rapmap_pseudoalign(fq1, fq2, rapmap_dir, gtf_file, fasta_file, data)
data = dd.set_transcriptome_bam(data, out_file)
return [[data]]
def _update_data(align_file, out_dir, names, data):
data = dd.set_work_bam(data, align_file)
data = dd.set_align_bam(data, align_file)
transcriptome_file = _move_transcriptome_file(out_dir, names)
data = dd.set_transcriptome_bam(data, transcriptome_file)
return data
def _update_data(align_file, out_dir, names, data):
data = dd.set_work_bam(data, align_file)
data = dd.set_align_bam(data, align_file)
transcriptome_file = _move_transcriptome_file(out_dir, names)
data = dd.set_transcriptome_bam(data, transcriptome_file)
return data
