def post_aln_qc(args, bam_file, logger=None):
""" perform post alignment quality check """
post_aln_dir = os.path.join(args.workDir, 'post_alignment_qc')
if not os.path.isdir(post_aln_dir):
os.mkdir(post_aln_dir)
#Fix mate information for bam file
exit_code, fix_mate_out = post_alignment_qc.fix_mate_information(args.picard, bam_file,
args.id, args.workDir, logger)
if exit_code == 0:
os.remove(bam_file)
assert(not os.path.isfile(bam_file))
os.rename(fix_mate_out, bam_file)
assert(os.path.isfile(bam_file))
#validate the post-alignment BAM file
post_alignment_qc.validate_bam_file(args.picard, bam_file, args.id, post_aln_dir, logger)
#collect RNA-seq metrics
post_alignment_qc.collect_rna_seq_metrics(args.picard, bam_file, args.id,
post_aln_dir, args.ref_flat, logger)
#run rna_seq_qc from broad institute
post_alignment_qc.bam_index(bam_file, args.id, logger)
rna_seq_qc_dir = os.path.join(post_aln_dir, 'rna_seq_qc')
if not os.path.isdir(rna_seq_qc_dir):
os.mkdir(rna_seq_qc_dir)
exit_code = post_alignment_qc.rna_seq_qc(args.rna_seq_qc_path, bam_file, args.id, rna_seq_qc_dir,
args.ref_genome,args.rna_seq_qc_annotation, logger)
if not(exit_code == 0):
reordered_bam = post_alignment_qc.reorder_bam(args.picard, bam_file, args.id, args.workDir,
args.ref_genome, logger)
post_alignment_qc.bam_index(reordered_bam, args.id, logger)
post_alignment_qc.rna_seq_qc(args.rna_seq_qc_path, reordered_bam, args.id, rna_seq_qc_dir,
args.ref_genome,args.rna_seq_qc_annotation, logger)
if os.path.isfile(reordered_bam):
os.remove(reordered_bam)
if os.path.isfile('%s.bai' %reordered_bam):
os.remove('%s.bai' %reordered_bam)
def post_aln_qc(args, bam_file, logger=None):
""" perform post alignment quality check """
#validate the post-alignment BAM file
post_alignment_qc.validate_bam_file(args.picard, bam_file, args.id, args.workDir, logger)
#collect RNA-seq metrics
post_alignment_qc.collect_rna_seq_metrics(args.picard, bam_file, args.id,
args.workDir, args.ref_flat, logger)
#run rna_seq_qc from broad institute
post_alignment_qc.bam_index(bam_file, args.id, logger)
exit_code = post_alignment_qc.rna_seq_qc(args.rna_seq_qc_path, bam_file, args.id, args.workDir,
args.ref_genome,args.rna_seq_qc_annotation, logger)
if not(exit_code == 0):
reordered_bam = post_alignment_qc.reorder_bam(args.picard, bam_file, args.id, args.workDir,
args.ref_genome, logger)
post_alignment_qc.bam_index(reordered_bam, args.id, logger)
post_alignment_qc.rna_seq_qc(args.rna_seq_qc_path, reordered_bam, args.id, args.workDir,
args.ref_genome,args.rna_seq_qc_annotation, logger)
def run_pipeline(args, workdir, analysis_id, fastq_dir, logger):
""" align datasets using STAR and compute expression using cufflinks """
tar_file_in = args.input_file
qc_dir = os.path.join(workdir, 'qc')
if not os.path.isdir(qc_dir):
os.mkdir(qc_dir)
decompress(tar_file_in, fastq_dir)
for fname in os.listdir(fastq_dir):
if fname.endswith("_1.fastq.gz") or fname.endswith("_1.fastq"):
reads_1 = os.path.join(fastq_dir, fname)
if fname.endswith("_2.fastq.gz") or fname.endswith("_2.fastq"):
reads_2 = os.path.join(fastq_dir, fname)
qc.fastqc(args.fastqc_path, reads_1, reads_2, qc_dir, analysis_id, logger)
star_output_dir = os.path.join(workdir, 'star_2_pass')
if os.path.isdir(star_output_dir):
pipelineUtil.remove_dir(star_output_dir)
os.mkdir(star_output_dir)
bam = "%s_star.bam" %os.path.join(star_output_dir, analysis_id)
if not os.path.isfile(bam):
star_cmd = ['time', '/usr/bin/time', 'python', args.star_pipeline,
'--genomeDir', args.genome_dir,
'--runThreadN', args.p,
'--tarFileIn', tar_file_in,
'--workDir', workdir,
'--out', bam,
'--genomeFastaFile', args.genome_fasta_file,
'--sjdbGTFfile', args.gtf
]
if args.quantMode != "":
star_cmd.append('--quantMode')
star_cmd.append(args.quantMode)
pipelineUtil.log_function_time("STAR", analysis_id, star_cmd, logger)
exit_code = 1
#Fix mate information for BAM
exit_code, fix_mate_out = post_alignment_qc.fix_mate_information(args.picard, bam,
analysis_id, workdir, logger)
if exit_code == 0:
os.remove(bam)
assert(not os.path.isfile(bam))
os.rename(fix_mate_out, bam)
assert(os.path.isfile(bam))
#validate the post alignment BAM file
post_alignment_qc.validate_bam_file(args.picard, bam, analysis_id, qc_dir, logger)
#collect RNA-seq metrics
post_alignment_qc.collect_rna_seq_metrics(args.picard, bam, analysis_id,
qc_dir, args.ref_flat, logger)
#quantify using cufflinks
cufflinks_cmd = ['time', '/usr/bin/time', 'python', args.cufflinks_pipeline,
'--bam', bam,
'--gtf', args.gtf,
'--analysis_id', analysis_id,
'--out', star_output_dir,
'--p', args.p,
'--multi_read_correct', 'True'
]
pipelineUtil.log_function_time("CUFFLINKS", analysis_id, cufflinks_cmd, logger)
