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:
dd.set_oncofuse_file(data, oncofuse_file)
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 detect_fusions(data):
data = to_single_data(data)
# support the old style of fusion mode calling
if dd.get_fusion_mode(data, False):
data = dd.set_fusion_caller(data, ["oncofuse", "pizzly"])
logger.warning(
"``fusion_mode`` is deprecated in favor of turning on "
"callers with ``fusion_caller``. It will run pizzly and "
"oncofuse for now, but will eventually have support "
"dropped.")
fusion_caller = dd.get_fusion_caller(data, [])
if "oncofuse" in fusion_caller:
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
if "pizzly" in fusion_caller:
pizzly_dir = pizzly.run_pizzly(data)
if pizzly_dir:
data = dd.set_pizzly_dir(data, pizzly_dir)
data["fusion"] = {
"fasta":
os.path.join(pizzly_dir,
"%s.fusions.fasta" % dd.get_sample_name(data)),
"json":
os.path.join(pizzly_dir, "%s.json" % dd.get_sample_name(data))
}
if "ericscript" in fusion_caller:
ericscript_dir = ericscript.run(data)
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 = dd.set_oncofuse_file(data, oncofuse_file)
if dd.get_transcriptome_align(data) and not dd.get_transcriptome_bam(data):
file1, file2 = None, None
if dd.get_disambiguate(data):
bam_path = data["work_bam"]
fastq_paths = alignprep._bgzip_from_bam(bam_path, data["dirs"], data["config"], is_retry=False, output_infix='-transcriptome')
if len(fastq_paths) == 2:
file1, file2 = fastq_paths
else:
file1, file2 = fastq_paths[0], None
else:
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("Transcriptome alignment was flagged to run, but the "
"transcriptome BAM file was not found. Aligning to the "
"transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
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 = dd.set_oncofuse_file(data, oncofuse_file)
if dd.get_transcriptome_align(data) and not dd.get_transcriptome_bam(data):
file1, file2 = None, None
if dd.get_disambiguate(data):
bam_path = data["work_bam"]
fastq_paths = alignprep._bgzip_from_bam(
bam_path,
data["dirs"],
data["config"],
is_retry=False,
output_infix='-transcriptome')
if len(fastq_paths) == 2:
file1, file2 = fastq_paths
else:
file1, file2 = fastq_paths[0], None
else:
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("Transcriptome alignment was flagged to run, but the "
"transcriptome BAM file was not found. Aligning to the "
"transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def detect_fusions(data):
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
pizzly_dir = pizzly.run_pizzly(data)
if pizzly_dir:
data = dd.set_pizzly_dir(data, pizzly_dir)
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 = dd.set_oncofuse_file(data, oncofuse_file)
# if RSEM set to run, but the aligner didn't create the transcriptome BAM
# file, make one with bwa
if dd.get_rsem(data) and not dd.get_transcriptome_bam(data):
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("RSEM was flagged to run, but the transcriptome BAM file "
"was not found. Aligning to the transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
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 = dd.set_oncofuse_file(data, oncofuse_file)
# if RSEM set to run, but the aligner didn't create the transcriptome BAM
# file, make one with bwa
if dd.get_rsem(data) and not dd.get_transcriptome_bam(data):
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info(
"RSEM was flagged to run, but the transcriptome BAM file "
"was not found. Aligning to the transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def detect_fusions(data):
# support the old style of fusion mode calling
if dd.get_fusion_mode(data, False):
data = dd.set_fusion_caller(data, ["oncofuse", "pizzly"])
logger.warning("``fusion_mode`` is deprecated in favor of turning on "
"callers with ``fusion_caller``. It will run pizzly and "
"oncofuse for now, but will eventually have support "
"dropped.")
if "oncofuse" in dd.get_fusion_caller(data, []):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
if "pizzly" in dd.get_fusion_caller(data, []):
pizzly_dir = pizzly.run_pizzly(data)
if pizzly_dir:
data = dd.set_pizzly_dir(data, pizzly_dir)
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 = dd.set_oncofuse_file(data, oncofuse_file)
# if RSEM set to run, but the aligner didn't create the transcriptome BAM
# file, make one with bwa
if dd.get_disambiguate(data):
logger.info("RSEM is not supported yet for disambiguation protocols. "
"See https://github.com/chapmanb/bcbio-nextgen/issues/859")
return [[data]]
if dd.get_rsem(data) and not dd.get_transcriptome_bam(data):
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("RSEM was flagged to run, but the transcriptome BAM file "
"was not found. Aligning to the transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def detect_fusions(data):
data = to_single_data(data)
# support the old style of fusion mode calling
if dd.get_fusion_mode(data, False):
data = dd.set_fusion_caller(data, ["oncofuse", "pizzly"])
logger.warning("``fusion_mode`` is deprecated in favor of turning on "
"callers with ``fusion_caller``. It will run pizzly and "
"oncofuse for now, but will eventually have support "
"dropped.")
fusion_caller = dd.get_fusion_caller(data, [])
if "oncofuse" in fusion_caller:
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
if "pizzly" in fusion_caller:
pizzly_dir = pizzly.run_pizzly(data)
if pizzly_dir:
data = dd.set_pizzly_dir(data, pizzly_dir)
data["fusion"] = {"fasta": os.path.join(pizzly_dir, "%s.fusions.fasta" % dd.get_sample_name(data)),
"json": os.path.join(pizzly_dir, "%s.json" % dd.get_sample_name(data))}
if "ericscript" in fusion_caller:
ericscript_dir = ericscript.run(data)
return [[data]]
