def process_alleles(vcf_record_in_one_contig, sample_names, curr_reference):
sample_to_allele = generate_empty_hash_with_sample(sample_names)
command = "samtools view -F 1028 %s %s" % (bam_file, curr_reference)
stream, process = utils_commands.get_output_stream_from_command(command)
for line in stream:
sam_record = Sam_record(line)
allele_array = []
sequence = sam_record.get_query_sequence()
sample = sam_record.get_tag("RG")
for position in vcf_record_in_one_contig.keys():
#if vcf_record_in_one_contig.get(position).get_genotype_quality(sample)>20:
allele_array.append(sequence[position - 1])
#else:
# allele_array.append('.')
count_with_hash(sample_to_allele[sample], ''.join(allele_array))
count_with_hash(sample_to_allele['all'], ''.join(allele_array))
process.wait()
pprint.pprint(sample_to_allele)
filter_alleles(sample_to_allele)
pprint.pprint(sample_to_allele)
all_alleles = set()
valid = True
for sample in sample_to_allele.keys():
alleles = sample_to_allele.get(sample)
all_alleles.update(set(alleles.keys()))
if len(alleles) > 2:
valid = False
if len(all_alleles) > 4:
valid = False
if not valid:
print curr_reference
def process_alleles(vcf_record_in_one_contig, sample_names,curr_reference):
sample_to_allele = generate_empty_hash_with_sample(sample_names)
command = "samtools view -F 1028 %s %s"%(bam_file,curr_reference)
stream,process=utils_commands.get_output_stream_from_command(command)
for line in stream:
sam_record=Sam_record(line)
allele_array=[]
sequence = sam_record.get_query_sequence()
sample = sam_record.get_tag("RG")
for position in vcf_record_in_one_contig.keys():
#if vcf_record_in_one_contig.get(position).get_genotype_quality(sample)>20:
allele_array.append(sequence[position-1])
#else:
# allele_array.append('.')
count_with_hash(sample_to_allele[sample], ''.join(allele_array))
count_with_hash(sample_to_allele['all'], ''.join(allele_array))
process.wait()
pprint.pprint(sample_to_allele)
filter_alleles(sample_to_allele)
pprint.pprint(sample_to_allele)
all_alleles=set()
valid=True
for sample in sample_to_allele.keys():
alleles = sample_to_allele.get(sample)
all_alleles.update(set(alleles.keys()))
if len(alleles)>2:
valid=False
if len(all_alleles)>4:
valid=False
if not valid:
print curr_reference
def process_single_samtools_run_with_read_group(bam_file, all_contigs_info, samtools_bin):
command = "%s view -h -F 132 %s" % (samtools_bin, bam_file)
open_stream, process = get_output_stream_from_command(command)
current_contig = None
sample_name, ext = os.path.splitext(bam_file)
read_groups = {}
try:
for line in open_stream:
if not line.startswith("@"):
break
if line.startswith("@RG"):
sp_line = line.strip().split()
rg_id = rg_sample = rg_library = None
for value in sp_line:
if value.startswith("ID"):
rg_id = value[3:]
elif value.startswith("SM"):
rg_sample = value[3:]
elif value.startswith("LB"):
rg_library = value[3:]
if rg_id:
if rg_sample:
read_groups[rg_id] = rg_sample
elif rg_library:
read_groups[rg_id] = rg_library
else:
read_groups[rg_id] = rg_id
all_sample_coverage = {}
all_sample_duplicate = {}
for sample in read_groups.values():
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
# process the first read
# if line.startswith("@"):
# #Still in the header. There's no read, exit
# return
sam_record = Sam_record(line.strip())
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
i = 1
# process all the others
for line in open_stream:
i += 1
if i % 1000000 == 0:
print i
sam_record = Sam_record(line.strip())
if current_contig != sam_record.get_reference_name() and current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(current_contig, all_sample_coverage.get(sample),
all_sample_duplicate.get(sample), sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
if current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(current_contig, all_sample_coverage.get(sample),
all_sample_duplicate.get(sample), sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
finally:
open_stream.close()
def process_single_samtools_run_with_read_group(bam_file, all_contigs_info,
samtools_bin):
command = "%s view -h -F 132 %s" % (samtools_bin, bam_file)
open_stream, process = get_output_stream_from_command(command)
current_contig = None
sample_name, ext = os.path.splitext(bam_file)
read_groups = {}
try:
for line in open_stream:
if not line.startswith("@"):
break
if line.startswith("@RG"):
sp_line = line.strip().split()
rg_id = rg_sample = rg_library = None
for value in sp_line:
if value.startswith("ID"):
rg_id = value[3:]
elif value.startswith("SM"):
rg_sample = value[3:]
elif value.startswith("LB"):
rg_library = value[3:]
if rg_id:
if rg_sample:
read_groups[rg_id] = rg_sample
elif rg_library:
read_groups[rg_id] = rg_library
else:
read_groups[rg_id] = rg_id
all_sample_coverage = {}
all_sample_duplicate = {}
for sample in read_groups.values():
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
# process the first read
# if line.startswith("@"):
# #Still in the header. There's no read, exit
# return
sam_record = Sam_record(line.strip())
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
i = 1
# process all the others
for line in open_stream:
i += 1
if i % 1000000 == 0:
print i
sam_record = Sam_record(line.strip())
if current_contig != sam_record.get_reference_name(
) and current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(
current_contig,
all_sample_coverage.get(sample),
all_sample_duplicate.get(sample),
sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
if current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(current_contig,
all_sample_coverage.get(sample),
all_sample_duplicate.get(sample),
sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
finally:
open_stream.close()
def process_single_samtools_run_with_read_group(bam_file,all_contigs_info,samtools_bin):
command="%s view -h -F 132 %s"%(samtools_bin, bam_file)
open_stream, process = get_output_stream_from_command(command)
current_contig=None
sample_name, ext = os.path.splitext(bam_file)
read_groups={}
try:
for line in open_stream:
if not line.startswith("@"):
break
if line.startswith("@RG"):
sp_line = line.strip().split()
rg_id=rg_sample=rg_library=None
for value in sp_line:
if value.startswith("ID"):
rg_id=value[3:]
elif value.startswith("SM"):
rg_sample=value[3:]
elif value.startswith("LB"):
rg_library=value[3:]
if rg_id:
if rg_sample:
read_groups[rg_id]=rg_sample
elif rg_library:
read_groups[rg_id]=rg_library
else:
read_groups[rg_id]=rg_id
all_sample_coverage={}
all_sample_coverage_reads = {}
all_sample_duplicate={}
for sample in read_groups.values():
all_sample_coverage[sample]=Counter()
all_sample_duplicate[sample]=Counter()
all_sample_coverage_reads[sample] = defaultdict(Counter)
#process the first read
sam_record = Sam_record(line.strip())
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
read_sequence = sam_record.get_query_sequence()
loci = get_loci_from_read(sam_record)
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage_reads[read_groups.get(rg_id)][str(loci)][read_sequence] +=1
i=1
#process all the others
for line in open_stream:
i+=1
if i%1000000==0:
print i
sam_record = Sam_record(line.strip())
if current_contig != sam_record.get_reference_name() and current_contig != None:
for sample in read_groups.values():
for loci in all_sample_coverage.get(sample):
alleles = all_sample_coverage_reads[sample].get(loci)
all_contigs_info.add_values(current_contig, loci, all_sample_coverage.get(sample).get(loci, 0),
all_sample_duplicate.get(sample).get(loci, 0), alleles=alleles,
sample=sample)
all_sample_coverage[sample]=Counter()
all_sample_duplicate[sample]=Counter()
all_sample_coverage_reads[sample] = defaultdict(Counter)
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
loci = get_loci_from_read(sam_record)
read_sequence = sam_record.get_query_sequence()
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage_reads[read_groups.get(rg_id)][str(loci)][read_sequence] +=1
if current_contig != None:
for sample in read_groups.values():
for loci in all_sample_coverage.get(sample):
alleles = all_sample_coverage_reads[sample].get(loci)
all_contigs_info.add_values(current_contig, loci, all_sample_coverage.get(sample).get(loci, 0),
all_sample_duplicate.get(sample).get(loci, 0), alleles=alleles,
sample=sample)
all_sample_coverage[sample]=Counter()
all_sample_duplicate[sample]=Counter()
all_sample_coverage_reads[sample] = defaultdict(Counter)
finally:
open_stream.close()
