def create_unaligned_bam(args, outs):
star_ref_path = cr_utils.get_reference_star_path(args.reference_path)
header_buf = cStringIO.StringIO()
header_buf.write('@HD\tVN:1.4\n')
# SQ header lines
with open(os.path.join(star_ref_path, 'chrNameLength.txt')) as f:
for line in f:
chr_name, chr_len = line.strip().split('\t')
header_buf.write('@SQ\tSN:{}\tLN:{}\n'.format(chr_name, chr_len))
# RG header lines
for packed_rg in args.read_groups:
header_buf.write(
re.sub('\\\\t', '\t', tk_bam.make_rg_header(packed_rg)) + '\n')
# Get read group ID for this chunk of reads
read_group = args.read_group
# pysam doesn't support reading SAM from a StringIO object
with open('tmphdr', 'w') as f:
f.write(header_buf.getvalue())
samfile = pysam.AlignmentFile('tmphdr', 'r', check_sq=False)
outbam = pysam.AlignmentFile(outs.genome_output, 'wb', template=samfile)
fastq_file1 = cr_io.open_maybe_gzip(args.read_chunk)
fastq_file2 = cr_io.open_maybe_gzip(
args.read2_chunk) if args.read2_chunk else None
read1s = tk_fasta.read_generator_fastq(fastq_file1)
read2s = tk_fasta.read_generator_fastq(fastq_file2) if fastq_file2 else []
record = pysam.AlignedSegment()
record.flag = 4
for read1, read2 in itertools.izip_longest(read1s, read2s):
name, seq, qual = read1
record.query_name, record.query_sequence = name.split(' ')[0], seq
record.query_qualities = tk_fasta.get_qvs(qual)
record.set_tag('RG', read_group, 'Z')
outbam.write(record)
if read2:
name, seq, qual = read2
record.query_name, record.query_sequence = name.split(' ')[0], seq
record.query_qualities = tk_fasta.get_qvs(qual)
record.set_tag('RG', read_group, 'Z')
outbam.write(record)
samfile.close()
fastq_file1.close()
if fastq_file2 is not None:
fastq_file2.close()
outbam.close()
def main(args, outs):
reference_star_path = cr_utils.get_reference_star_path(args.reference_path)
star = cr_reference.STAR(reference_star_path)
star.align(args.read_chunk,
args.read2_chunk,
outs.genome_output,
max_report_alignments_per_read=args.max_hits_per_read,
threads=args.threads,
read_group_tags=tk_bam.make_star_rg_header(args.read_group))
def main(args, outs):
reference_star_path = cr_utils.get_reference_star_path(args.reference_path)
star_index = cr_transcriptome.build_star_index(reference_star_path)
chroms = star_index[0][0]
gene_index = cr_reference.GeneIndex.load_pickle(cr_utils.get_reference_genes_index(args.reference_path))
barcode_whitelist = cr_utils.load_barcode_whitelist(args.barcode_whitelist)
barcode_dist = cr_utils.load_barcode_dist(args.barcode_counts, barcode_whitelist, args.gem_group)
reporter = cr_report.Reporter(reference_path=args.reference_path,
high_conf_mapq=cr_constants.STAR_DEFAULT_HIGH_CONF_MAPQ,
gene_index=gene_index,
chroms=chroms,
barcode_whitelist=barcode_whitelist,
barcode_dist=barcode_dist,
gem_groups=args.gem_groups,
umi_length=cr_chem.get_umi_length(args.chemistry_def),
umi_min_qual_threshold=args.umi_min_qual_threshold)
reporter.attach_bcs_init()
outs.num_alignments = process_alignments(args.chunk_genome_input, args.chunk_trimmed_input, outs.output, args.bam_comments, reporter, gene_index, star_index, args)
reporter.attach_bcs_finalize()
reporter.save(outs.chunked_reporter)