def write_sub_bam(chrom_list, used_bam_file_tmp, exclude_bam_file_tmp, out_dir,
total_modify_reads_file, total_delete_reads_file,
total_add_reads_file, process):
write_bam_pool = Pool(int(process))
exclude_bam_list = [exclude_bam_file_tmp]
usedBamList = []
for chrom in chrom_list:
excludeBam_chr = "%s/exclude_%s.bam" % (out_dir, chrom)
exclude_bam_list.append(excludeBam_chr)
usedBam_chr = "%s/used_%s.bam" % (out_dir, chrom)
usedBamList.append(usedBam_chr)
write_bam_pool.apply_async(
write_bam_byChr,
args=(used_bam_file_tmp, chrom, excludeBam_chr, usedBam_chr,
total_modify_reads_file, total_delete_reads_file,
total_add_reads_file))
write_bam_pool.close()
write_bam_pool.join()
exclude_bam_file = os.path.join(out_dir, "exclude.bam")
bamMerge(exclude_bam_list, exclude_bam_file)
used_bam_file = os.path.join(out_dir, "used.bam")
if len(usedBamList) != 1:
bamMerge(usedBamList, used_bam_file)
else:
used_bam_file = usedBamList[0]
bamSort(used_bam_file, os.path.join(out_dir, "used.sort"))
used_sort_bam_file = os.path.join(out_dir, "used.sort.bam")
bamIndex(used_sort_bam_file)
return used_sort_bam_file, exclude_bam_file
def get_reads_by_region(bam_file, sv_list, out_dir):
# get reads by region bed
region_bed_file = os.path.join(out_dir, "consider_region.bed")
chrom_list = write_region_bed(region_bed_file, sv_list)
exclude_bam_file_tmp = os.path.join(out_dir, "exclude_tmp.bam")
used_bam_file_tmp = os.path.join(out_dir, "used_tmp.bam")
getRegionReads(bam_file, region_bed_file, used_bam_file_tmp,
exclude_bam_file_tmp)
bamIndex(used_bam_file_tmp)
return chrom_list, used_bam_file_tmp, exclude_bam_file_tmp
def main(run_args):
start_time = time.asctime(time.localtime(time.time()))
# print start_time
temp_out_dir = os.path.join(run_args.outdir, "tempDir")
os.system("mkdir -p %s" % temp_out_dir)
invalid_log_file = os.path.join(run_args.outdir, 'invalid_mutation.txt')
invalid_log = InvalidLog(invalid_log_file)
# step1: deal with mutfile and get haplotypes
print "step1: deal with mutfile and get haplotypes"
haplotype_list = get_haplotypes(run_args.bamfile, run_args.reffasta,
run_args.mutfile, int(run_args.haplosize),
float(run_args.snpfrac), invalid_log)
# step2: deal haplotypes and get total_chosen_reads, total_chosen_reads_muts
print "step2: deal haplotypes and get total_chosen_reads, total_chosen_reads_muts"
success_list_file = os.path.join(run_args.outdir, 'success_list.txt')
total_chosen_reads, total_chosen_reads_muts = deal_haplotype_multi(
run_args.bamfile, haplotype_list, temp_out_dir, run_args.reffasta,
int(run_args.process), int(run_args.mindepth),
int(run_args.minmutreads), int(run_args.minmapq),
float(run_args.diffcover), run_args.single, run_args.multmapfilter,
run_args.aligner, run_args.alignerIndex, invalid_log,
success_list_file)
invalid_log.close()
if len(total_chosen_reads) == 0:
print "Warning: No reads to deal with of all these sv, checkout your sv file"
return
# step3: modify the reads in total_chosen_reads itself
print "step3: modify the reads in total_chosen_reads itself"
reads_modify(total_chosen_reads, total_chosen_reads_muts,
run_args.reffasta, int(run_args.process))
# step4: write edited reads to edited file and exclude reads to exclude file ,than remap edited file to reference
print "step4: write edited reads to edited file and exclude reads to exclude file ,than remap edited file to reference"
edit_remap_bam_file, exclude_bam_file = reads_replace(
run_args.bamfile, total_chosen_reads, run_args.seqer,
run_args.floworder, run_args.libkey, run_args.barcode, run_args.tag,
temp_out_dir, run_args.aligner, run_args.alignerIndex, run_args.single)
# step5: merge remap.edit.bam and exclude exclude.bam and sort
print "step5: merge remap.edit.bam and exclude exclude.bam and sort"
# edit_remap_bam_file, exclude_bam_file = os.path.join(temp_out_dir, "edit.remap.sort.bam"), os.path.join(
# temp_out_dir, "exclude.bam")
out_bam_file = os.path.join(run_args.outdir, "edit.sorted.bam")
bamMerge([edit_remap_bam_file, exclude_bam_file], out_bam_file)
bamIndex(out_bam_file)
end_time = time.asctime(time.localtime(time.time()))
# speed_time = end_time - start_time
print "Edit Bam is completed! Result see %s and valid mutation see %s. Invalid mutation can't be spike in see %s." % (
out_bam_file, success_list_file, invalid_log_file)
def reads_replace(bam_file, total_chosen_reads, seqer, flow_order, lib_key,
barcode, tag, out_dir, aligner, aligner_index, is_single):
bam = pysam.AlignmentFile(bam_file)
edit_bam_reads = {}
for read in bam.fetch():
read_name = read.query_name
if read_name in total_chosen_reads:
strand = getReadStrand(read)
if read_name not in edit_bam_reads:
edit_bam_reads[read_name] = {}
if strand in total_chosen_reads[read_name]:
my_read = total_chosen_reads[read_name][strand]
read.query_sequence = my_read.query_sequence
read.query_qualities = my_read.query_qualities
if seqer == "life":
read = deal_life_reads(read, flow_order, lib_key, barcode)
if tag:
read = add_tag(read)
edit_bam_reads[read_name][strand] = read
else:
edit_bam_reads[read_name][strand] = read
# write edited reads into edit.bam
edit_bam_file = os.path.join(out_dir, "edit.bam")
edit_bam = pysam.AlignmentFile(edit_bam_file, 'wb', template=bam)
for read_name, readInfo in edit_bam_reads.items():
for strand, read in readInfo.items():
edit_bam.write(read)
edit_bam.close()
# write not edited reads into exclude.bam
exclude_bam_file = os.path.join(out_dir, "exclude.bam")
exclude_bam = pysam.AlignmentFile(exclude_bam_file, 'wb', template=bam)
for read in bam.fetch():
read_name = read.query_name
if read_name not in edit_bam_reads:
exclude_bam.write(read)
exclude_bam.close()
# remap the edited reads
header = os.path.join(out_dir, 'bam.header')
os.system('samtools view -H %s|grep "^@RG" > %s' % (bam_file, header))
head = open(header, 'r').readline().rstrip()
if not head:
head = None
edit_remap_bam_file = os.path.join(out_dir, "edit.remap.bam")
remap(aligner_index,
edit_bam_file,
edit_remap_bam_file,
aligner,
is_single,
header=head)
edit_remap_bam_sorted_prefix = os.path.join(out_dir, "edit.remap.sort")
edit_remap_bam_sorted_file = os.path.join(out_dir, "edit.remap.sort.bam")
bamSort(edit_remap_bam_file, edit_remap_bam_sorted_prefix)
bamIndex(edit_remap_bam_sorted_file)
if tag:
edit_remap_addtag_file = os.path.join(out_dir, "edit.remap.sort.bam")
bam_add_tag(edit_remap_bam_sorted_file, edit_remap_addtag_file)
else:
edit_remap_addtag_file = edit_remap_bam_sorted_file
return edit_remap_addtag_file, exclude_bam_file
def main(run_args):
start_time = time.asctime(time.localtime(time.time()))
# print start_time
if not os.path.exists(run_args.outdir):
os.mkdir(run_args.outdir)
invalid_log_file = os.path.join(run_args.outdir, 'invalid_mutation.txt')
invalid_log = InvalidLog(invalid_log_file)
run_log_file = os.path.join(run_args.outdir, 'run.log')
run_log = RunLog(run_log_file)
temp_out_dir = os.path.join(run_args.outdir, "tempDir")
if not os.path.exists(temp_out_dir):
os.mkdir(temp_out_dir)
# step0: prepare sv list
sv_list = check_sv_file(run_args.svfile, run_args.reffasta, invalid_log)
if not sv_list:
exit("no sv list to deal with")
# step1: get insert size of paired reads
print "step1: get insert size of paired reads"
insert_size = get_insertSize_range(run_args.bamfile, run_args.readlength, run_args.single)
# step2: deal with sv and get total edited reads
print "step2: deal with sv and get total edited reads"
success_file = os.path.join(run_args.outdir, 'success_list.txt')
total_modify_reads_file, total_delete_reads_file, total_add_reads_file, total_modify_list, total_delete_list, total_add_list = deal_sv(
run_args.bamfile, run_args.reffasta, sv_list,
run_args.single,
int(run_args.minmapq),
run_args.multmapfilter,
int(run_args.mindepth),
int(run_args.minmutreads),
int(run_args.readlength),
temp_out_dir,
insert_size,
invalid_log,
run_log, success_file)
invalid_log.close()
# step3: get reads by region bed and write bam file
print "step3: get reads by region bed and write bam file"
chrom_list, used_bam_file_tmp, exclude_bam_file_tmp = get_reads_by_region(run_args.bamfile, sv_list,
temp_out_dir)
# write reads which may probably used to used.bam and reads should not be used to exclude.bam
used_bam_file, exclude_bam_file = write_sub_bam(chrom_list, used_bam_file_tmp, exclude_bam_file_tmp,
temp_out_dir, total_modify_reads_file,
total_delete_reads_file, total_add_reads_file,
int(run_args.process))
# step4: merge edited reads and remap to new bam, consider about the tag, RG, life reads
print "step4: merge edited reads and remap to new bam, consider about the tag, RG, life reads"
edit_remap_bam_file = merge_edit_bam(run_args.bamfile, temp_out_dir, run_args.single, total_modify_reads_file,
total_add_reads_file, used_bam_file, total_modify_list, total_add_list,
run_args.seqer, run_args.aligner, run_args.alignerIndex,
run_args.floworder,
run_args.libkey, run_args.barcode, run_args.tag)
# step5: remapped edit reads and merge
print "step5: remapped edit reads and merge"
out_bam_file = os.path.join(run_args.outdir, "edit.sorted.bam")
bamMerge([edit_remap_bam_file, exclude_bam_file], out_bam_file)
bamIndex(out_bam_file)
end_time = time.asctime(time.localtime(time.time()))
# print end_time
# speed_time = end_time - start_time
print "Edit Bam is completed! Result see %s and valid mutation see %s. Invalid mutation can't be spike in see %s." % (
out_bam_file, success_file, invalid_log_file)