def align(fastq_file, pair_file, ref_file, names, align_dir, data): config = data["config"] out_prefix = os.path.join(align_dir, dd.get_lane(data)) out_file = out_prefix + "Aligned.out.sam" out_dir = os.path.join(align_dir, "%s_star" % dd.get_lane(data)) if not ref_file: logger.error("STAR index not found. We don't provide the STAR indexes " "by default because they are very large. You can install " "the index for your genome with: bcbio_nextgen.py upgrade " "--aligners star --genomes genome-build-name --data") sys.exit(1) final_out = os.path.join(out_dir, "{0}.bam".format(names["sample"])) if file_exists(final_out): return final_out star_path = config_utils.get_program("STAR", config) fastq = " ".join([fastq_file, pair_file]) if pair_file else fastq_file num_cores = config["algorithm"].get("num_cores", 1) safe_makedir(align_dir) cmd = ("{star_path} --genomeDir {ref_file} --readFilesIn {fastq} " "--runThreadN {num_cores} --outFileNamePrefix {out_prefix} " "--outReadsUnmapped Fastx --outFilterMultimapNmax 10 " "--outStd SAM " "--outSAMunmapped Within --outSAMattributes %s" % " ".join(ALIGN_TAGS)) cmd = cmd + " --readFilesCommand zcat " if is_gzipped(fastq_file) else cmd cmd += _read_group_option(names) fusion_mode = utils.get_in(data, ("config", "algorithm", "fusion_mode"), False) if fusion_mode: cmd += " --chimSegmentMin 15 --chimJunctionOverhangMin 15" strandedness = utils.get_in(data, ("config", "algorithm", "strandedness"), "unstranded").lower() if strandedness == "unstranded": cmd += " --outSAMstrandField intronMotif " if dd.get_rsem(data): cmd += " --quantMode TranscriptomeSAM " with tx_tmpdir(data) as tmp_dir: sam_to_bam = bam.sam_to_bam_stream_cmd(config) sort = bam.sort_cmd(config, tmp_dir) cmd += "| {sam_to_bam} | {sort} -o {tx_final_out} " run_message = "Running STAR aligner on %s and %s" % (fastq_file, ref_file) with file_transaction(data, final_out) as tx_final_out: do.run(cmd.format(**locals()), run_message, None) if dd.get_rsem(data): transcriptome_file = _move_transcriptome_file(out_dir, names) return final_out
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_rsem(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 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 = 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 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]]