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 main(run_args):
invalid_log_file = os.path.join(run_args.outdir, 'invalid_mutation.txt')
invalid_log = log(invalid_log_file)
temp_out_dir = os.path.join(run_args.out_dir, "tempDir")
os.system("mkdir -p %s" % temp_out_dir)
# 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
total_chosen_reads, total_chosen_reads_muts = deal_haplotype_multi(run_args.bam_file, 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.is_multmapfilter, run_args.aligner,
run_args.alignerIndex, invalid_log)
# step3: modify the reads in total_chosen_reads itself
reads_modify(total_chosen_reads, total_chosen_reads_muts, run_args.reffasta, int(run_args.prceoss))
# 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.bam_file, total_chosen_reads, run_args.seqer,
run_args.floworder, run_args.lib_key, run_args.barcode,
run_args.tag, temp_out_dir, run_args.aligner,
run_args.aligner_index, run_args.single)
# step5: merge remap.edit.bam and exclude exclude.bam and sort
bamIndex(exclude_bam_file)
out_bam_file = os.path.join(temp_out_dir, "edit_exclude.bam")
bamMerge([edit_remap_bam_file, exclude_bam_file], out_bam_file)
out_sort_bam_file = os.path.join(run_args.outdir, "edit.sort.bam")
out_sort_bam_file_prefix = os.path.join(run_args.outdir, "edit.sort")
bamSort(out_bam_file, out_sort_bam_file_prefix)
bamIndex(out_sort_bam_file)
print "Edit Bam is completed! Result see %s and invalid mutation can't be spike in see %s." % (
out_sort_bam_file, invalid_log_file)
def write_sub_bam(chrom_list, used_bam_file_tmp, exclude_bam_file_tmp, out_dir,
total_modify_readname_list, total_delete_readname_list,
total_add_readname_list, 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_readname_list, total_delete_readname_list,
total_add_readname_list))
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)
used_bam = pysam.AlignmentFile(used_sort_bam_file, 'rb')
used_reads = {}
for read in used_bam.fetch():
keyname = getKeyName(read)
used_reads[keyname] = read
used_bam.close()
return used_reads, used_bam_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 run_args.debug:
print len(sv_list)
if not sv_list:
exit("no sv list to deal with")
# step1: get insert size of paired reads
insert_size = get_insertSize_range(run_args.bamfile, run_args.readlength,
run_args.single, run_args.debug)
if run_args.debug:
print insert_size
# step2: deal with sv
total_modify_reads, total_delete_reads, total_add_reads = 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)
# exit
total_deal_reads_num = len(total_modify_reads) + len(
total_delete_reads) + len(total_add_reads)
if total_deal_reads_num == 0:
# run_log.info("No reads to deal with of all these sv, check out your sv file")
print "Warning: No reads to deal with of all these sv, checkout your sv file"
return
if run_args.debug:
print len(total_modify_reads), len(total_delete_reads), len(
total_add_reads)
# step3: merge edit reads
total_modify_readname_list, total_delete_readname_list, total_add_readname_list = merge_edit_reads(
total_modify_reads, total_add_reads, total_delete_reads)
if run_args.debug:
print "list num: ", len(total_modify_readname_list), len(
total_delete_readname_list), len(total_add_readname_list)
# step4: 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_reads, used_bam_file, exclude_bam_file = write_sub_bam(
chrom_list, used_bam_file_tmp, exclude_bam_file_tmp, temp_out_dir,
total_modify_readname_list, total_delete_readname_list,
total_add_readname_list, int(run_args.process))
if run_args.debug:
print "used & exclude bam:", used_bam_file, exclude_bam_file
# step5: 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,
total_add_reads, used_reads, run_args.seqer, run_args.aligner,
run_args.aligner_index, run_args.flow_order, run_args.lib_key,
run_args.barcode, run_args.tag)
if run_args.debug:
print "edit remap bam:", edit_remap_bam_file
# step6: read remapped edit reads to dict
out_bam_file = os.path.join(temp_out_dir, "edit_exclude.bam")
bamMerge([edit_remap_bam_file, exclude_bam_file], out_bam_file)
bamIndex(out_bam_file)
out_sort_bam_file = os.path.join(run_args.outdir, "edit.sort.bam")
out_sort_bam_file_prefix = os.path.join(run_args.outdir, "edit.sort")
bamSort(out_bam_file, out_sort_bam_file_prefix)
bamIndex(out_sort_bam_file)
end_time = time.asctime(time.localtime(time.time()))
print end_time
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)
def merge_edit_bam(bam_file, out_dir, is_single, total_modify_reads,
total_add_reads, used_reads, seqer, aligner, aligner_index,
flow_order, lib_key, barcode, tag):
bam = pysam.AlignmentFile(bam_file, 'rb')
edit_bam_file = os.path.join(out_dir, "edit.bam")
edit_bam = pysam.AlignmentFile(edit_bam_file, 'wb', template=bam)
readname_convert_file = os.path.join(out_dir, "readname_convert.txt")
fout_convert = open(readname_convert_file, 'w')
edit_bam_reads = {}
if is_single:
for read_pair in total_modify_reads:
read1 = read_pair[0]
keyname_read1 = getKeyName(read1)
orig_read1 = used_reads[keyname_read1]
new_read1 = copy.deepcopy(orig_read1)
new_read1.query_sequence = read1.query_sequence
new_read1.query_qualities = read1.query_qualities
new_name = read1.query_name.split(
":")[0] + ":" + get_new_readname()
new_read1.query_name = new_name
if seqer == "life":
new_read1 = deal_life_reads(new_read1, flow_order, lib_key,
barcode)
if tag:
new_read1 = add_tag(new_read1)
edit_bam.write(new_read1)
fout_convert.write(
"%s: %s, %s, %s-%s\n" %
(new_name, orig_read1.query_name, new_read1.is_read1,
new_read1.reference_start, new_read1.reference_end))
strand = getReadStrand(new_read1)
if new_name not in edit_bam_reads:
edit_bam_reads[new_name] = dict()
edit_bam_reads[new_name][strand] = new_read1
for read_pair in total_add_reads:
read1 = read_pair[0]
keyname_read1 = getKeyName(read1)
orig_read1 = used_reads[keyname_read1]
new_read1 = copy.deepcopy(orig_read1)
new_name = get_new_readname()
new_read1.query_name = new_name
edit_bam.write(new_read1)
fout_convert.write(
"%s: %s, %s, %s-%s\n" %
(new_name, orig_read1.query_name, new_read1.is_read1,
new_read1.reference_start, new_read1.reference_end))
strand = getReadStrand(new_read1)
if new_name not in edit_bam_reads:
edit_bam_reads[new_name] = dict()
edit_bam_reads[new_name][strand] = new_read1
else:
for read_pair in total_modify_reads + total_add_reads:
read1 = read_pair[0]
read2 = read_pair[1]
keyname_read1 = getKeyName(read1)
keyname_read2 = getKeyName(read2)
orig_read1 = used_reads[keyname_read1]
orig_read2 = used_reads[keyname_read2]
orig_read1_name = orig_read1.query_name
orig_read2_name = orig_read2.query_name
new_read1 = copy.deepcopy(orig_read1)
new_read2 = copy.deepcopy(orig_read2)
new_read1.query_sequence = read1.query_sequence
new_read1.query_qualities = read1.query_qualities
new_read2.query_sequence = read2.query_sequence
new_read2.query_qualities = read2.query_qualities
new_name = get_new_readname()
new_read1.query_name = new_name
new_read2.query_name = new_name
strand1 = getReadStrand(new_read1)
strand2 = getReadStrand(new_read2)
if new_name not in edit_bam_reads:
edit_bam_reads[new_name] = dict()
edit_bam_reads[new_name][strand1] = new_read1
edit_bam_reads[new_name][strand2] = new_read2
if tag:
new_read1 = add_tag(new_read1)
new_read2 = add_tag(new_read2)
fout_convert.write("%s: %s, %s, %s, %s, %s-%s, %s-%s\n" % (
new_name,
orig_read1_name,
orig_read2_name,
new_read1.is_read1,
new_read2.is_read2,
new_read1.reference_start,
new_read1.reference_end,
new_read2.reference_start,
new_read2.reference_end,
))
edit_bam.write(new_read1)
edit_bam.write(new_read2)
fout_convert.close()
edit_bam.close()
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)
if not is_single:
editRemap = pysam.AlignmentFile(edit_remap_bam_file, 'rb')
editRemapBam_addRG_File = os.path.join(out_dir, "edit.remap.addRG.bam")
bamAddRG(editRemap, edit_bam_reads, bam, editRemapBam_addRG_File)
editRemap.close()
else:
editRemapBam_addRG_File = edit_remap_bam_file
bamIndex(editRemapBam_addRG_File)
return editRemapBam_addRG_File
