def main():
# read params
args = parse_arguments()
REF = args.ref_fa
FINAL_BAM = args.nodup_bam
OUTPUT_PREFIX = os.path.join(
args.out_dir,
os.path.basename(strip_ext_bam(FINAL_BAM)))
RG_FREE_FINAL_BAM = remove_read_group(FINAL_BAM)
JAVA_HEAP = args.picard_java_heap
gc_out, gc_plot_pdf, gc_summary = get_gc(RG_FREE_FINAL_BAM,
REF,
OUTPUT_PREFIX,
JAVA_HEAP)
# will generate PNG format from gc_out
plot_gc(gc_out, OUTPUT_PREFIX)
rm_f(RG_FREE_FINAL_BAM)
log.info('List all files in output directory...')
ls_l(args.out_dir)
log.info('All done.')
def sambamba_name_sort(bam, nth, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
nmsrt_bam = '{}.nmsrt.bam'.format(prefix)
cmd = 'sambamba sort -n {} -o {} -t {}'.format(bam, nmsrt_bam, nth)
run_shell_cmd(cmd)
return nmsrt_bam
def main():
# read params
args = parse_arguments()
ALIGNED_BAM = args.bam
OUTPUT_PREFIX = os.path.join(args.out_dir,
os.path.basename(strip_ext_bam(ALIGNED_BAM)))
RG_FREE_ALIGNED_BAM = remove_read_group(ALIGNED_BAM)
JAVA_HEAP = args.picard_java_heap
# Library complexity: Preseq results, NRF, PBC1, PBC2
if args.paired_end:
picard_est_lib_size = get_picard_complexity_metrics(
RG_FREE_ALIGNED_BAM, OUTPUT_PREFIX, JAVA_HEAP)
else:
picard_est_lib_size = None
preseq_data, preseq_log = run_preseq(ALIGNED_BAM,
OUTPUT_PREFIX) # SORTED BAM
get_preseq_plot(preseq_data, OUTPUT_PREFIX)
# write picard_est_lib_size to file
if picard_est_lib_size is not None:
picard_est_lib_size_file = OUTPUT_PREFIX + '.picard_est_lib_size.qc'
with open(picard_est_lib_size_file, 'w') as fp:
fp.write(str(picard_est_lib_size) + '\n')
rm_f(RG_FREE_ALIGNED_BAM)
log.info('List all files in output directory...')
ls_l(args.out_dir)
log.info('All done.')
def sambamba_flagstat(bam, nth, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
flagstat_qc = '{}.flagstat.qc'.format(prefix)
cmd = 'sambamba flagstat {} -t {} > {}'.format(bam, nth, flagstat_qc)
run_shell_cmd(cmd)
return flagstat_qc
def remove_chrs_from_bam(bam, chrs, chrsz, nth=1, out_dir=''):
if len(chrs) == 0:
raise ValueError('There must be at least one chromosome, zero found.')
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
suffix = 'no_{}'.format('_'.join(chrs))
final_bam = '{}.{}.bam'.format(prefix, suffix)
tmp_chrsz = '{}.{}.tmp.chrsz'.format(prefix, suffix)
# make a temp chrsz file
cmd0 = 'zcat -f {chrsz} |'
cmd0 += 'grep -v -P \'^({chrs})\\s\' | '
cmd0 += 'awk \'BEGIN{{OFS="\\t"}} {{print $1,0,$2}}\' > {tmp_chrsz}'
cmd0 = cmd0.format(chrsz=chrsz, chrs='|'.join(chrs), tmp_chrsz=tmp_chrsz)
run_shell_cmd(cmd0)
# remove chrs from BAM
cmd1 = 'samtools view -b -L {tmp_chrsz} {bam} {res_param} > {final_bam}'
cmd1 = cmd1.format(tmp_chrsz=tmp_chrsz,
bam=bam,
res_param=get_samtools_res_param('view', nth=nth),
final_bam=final_bam)
run_shell_cmd(cmd1)
rm_f(tmp_chrsz)
return final_bam
def rm_unmapped_lowq_reads_se(bam, multimapping, mapq_thresh, nth, out_dir):
prefix = os.path.join(out_dir,
os.path.basename(strip_ext_bam(bam)))
filt_bam = '{}.filt.bam'.format(prefix)
if multimapping:
qname_sort_bam = samtools_name_sort(bam, nth, out_dir)
cmd2 = 'samtools view -h {} | '
cmd2 += '$(which assign_multimappers.py) -k {} | '
cmd2 += 'samtools view -F 1804 -Su /dev/stdin | '
cmd2 += 'samtools sort /dev/stdin -o {} -T {} -@ {}'
cmd2 = cmd2.format(
qname_sort_bam,
multimapping,
filt_bam,
prefix,
nth)
run_shell_cmd(cmd2)
rm_f(qname_sort_bam) # remove temporary files
else:
cmd = 'samtools view -F 1804 -q {} -u {} | '
cmd += 'samtools sort /dev/stdin -o {} -T {} -@ {}'
cmd = cmd.format(
mapq_thresh,
bam,
filt_bam,
prefix,
nth)
run_shell_cmd(cmd)
return filt_bam
def rm_unmapped_lowq_reads_pe(bam, multimapping, mapq_thresh, nth, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
filt_bam = '{}.filt.bam'.format(prefix)
tmp_filt_bam = '{}.tmp_filt.bam'.format(prefix)
fixmate_bam = '{}.fixmate.bam'.format(prefix)
if multimapping:
cmd1 = 'samtools view -F 524 -f 2 -u {} | '
cmd1 += 'samtools sort -n /dev/stdin -o {} -T {} -@ {} '
cmd1 = cmd1.format(bam, tmp_filt_bam, prefix, nth)
run_shell_cmd(cmd1)
cmd2 = 'samtools view -h {} -@ {} | '
cmd2 += '$(which assign_multimappers.py) -k {} --paired-end | '
cmd2 += 'samtools fixmate -r /dev/stdin {}'
cmd2 = cmd2.format(tmp_filt_bam, nth, multimapping, fixmate_bam)
run_shell_cmd(cmd2)
else:
cmd1 = 'samtools view -F 1804 -f 2 -q {} -u {} | '
cmd1 += 'samtools sort -n /dev/stdin -o {} -T {} -@ {}'
cmd1 = cmd1.format(mapq_thresh, bam, tmp_filt_bam, prefix, nth)
run_shell_cmd(cmd1)
cmd2 = 'samtools fixmate -r {} {}'
cmd2 = cmd2.format(tmp_filt_bam, fixmate_bam)
run_shell_cmd(cmd2)
rm_f(tmp_filt_bam)
cmd = 'samtools view -F 1804 -f 2 -u {} | '
cmd += 'samtools sort /dev/stdin -o {} -T {} -@ {}'
cmd = cmd.format(fixmate_bam, filt_bam, prefix, nth)
run_shell_cmd(cmd)
rm_f(fixmate_bam)
return filt_bam
def mark_dup_picard(bam, out_dir, java_heap=None): # shared by both se and pe
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'filt')
dupmark_bam = '{}.dupmark.bam'.format(prefix)
dup_qc = '{}.dup.qc'.format(prefix)
if java_heap is None:
java_heap_param = '-Xmx4G'
else:
java_heap_param = '-Xmx{}'.format(java_heap)
run_shell_cmd('java {java_heap_param} -XX:ParallelGCThreads=1 '
'-jar {picard} MarkDuplicates '
'INPUT={bam} '
'OUTPUT={dupmark_bam} '
'METRICS_FILE={dup_qc} '
'VALIDATION_STRINGENCY=LENIENT '
'USE_JDK_DEFLATER=TRUE '
'USE_JDK_INFLATER=TRUE '
'ASSUME_SORTED=TRUE '
'REMOVE_DUPLICATES=FALSE '.format(
java_heap_param=java_heap_param,
picard=locate_picard(),
bam=bam,
dupmark_bam=dupmark_bam,
dup_qc=dup_qc,
))
return dupmark_bam, dup_qc
def samtools_sort(bam, nth, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
srt_bam = '{}.srt.bam'.format(prefix)
cmd = 'samtools sort {} -o {} -T {} -@ {}'.format(bam, srt_bam, prefix,
nth)
run_shell_cmd(cmd)
return srt_bam
def samstat(bam, nth=1, out_dir=''):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
samstat_qc = '{}.samstats.qc'.format(prefix)
cmd = 'samtools sort -n {bam} -T {prefix}.tmp -O sam | '
cmd += 'SAMstats --sorted_sam_file - --outf {samstat_qc}'
cmd = cmd.format(bam=bam, prefix=prefix, samstat_qc=samstat_qc)
run_shell_cmd(cmd)
return samstat_qc
def bam2ta_se(bam, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
ta = '{}.tagAlign.gz'.format(prefix)
cmd = 'bedtools bamtobed -i {} | '
cmd += 'awk \'BEGIN{{OFS="\\t"}}{{$4="N";$5="1000";print $0}}\' | '
cmd += 'gzip -nc > {}'
cmd = cmd.format(bam, ta)
run_shell_cmd(cmd)
return ta
def rm_dup_pe(dupmark_bam, nth, out_dir):
prefix = os.path.join(out_dir,
os.path.basename(strip_ext_bam(dupmark_bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'dupmark')
nodup_bam = '{}.nodup.bam'.format(prefix)
cmd1 = 'samtools view -@ {} -F 1804 -f 2 -b {} > {}'
cmd1 = cmd1.format(nth, dupmark_bam, nodup_bam)
run_shell_cmd(cmd1)
return nodup_bam
def samtools_name_sort(bam, nth=1, mem_gb=None, out_dir=''):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
nmsrt_bam = '{}.nmsrt.bam'.format(prefix)
run_shell_cmd(
'samtools sort -n {bam} -o {nmsrt_bam} -T {prefix} {res_param}'.format(
bam=bam,
nmsrt_bam=nmsrt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort', nth=nth, mem_gb=mem_gb),
))
return nmsrt_bam
def remove_read_group(bam, out_dir='.'):
basename = os.path.basename(strip_ext_bam(bam))
prefix = os.path.join(out_dir, basename)
new_bam = '{}.no_rg.bam'.format(prefix)
cmd = 'samtools view -h {} | '
cmd += 'grep -v "^@RG" | sed "s/\\tRG:Z:[^\\t]*//" | '
cmd += 'samtools view -bo {} -'
cmd = cmd.format(bam, new_bam)
run_shell_cmd(cmd)
return new_bam
def samstat(bam, nth=1, mem_gb=None, out_dir=''):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
samstat_qc = '{}.samstats.qc'.format(prefix)
run_shell_cmd(
'samtools sort -n {bam} -T {prefix}.tmp {res_param} -O sam | '
'SAMstats --sorted_sam_file - --outf {samstat_qc}'.format(
bam=bam,
prefix=prefix,
res_param=get_samtools_res_param('sort', nth=nth, mem_gb=mem_gb),
samstat_qc=samstat_qc,
))
return samstat_qc
def mark_dup_sambamba(bam, nth, out_dir): # shared by both se and pe
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'filt')
dupmark_bam = '{}.dupmark.bam'.format(prefix)
dup_qc = '{}.dup.qc'
cmd = 'sambamba markdup -t {} --hash-table-size=17592186044416 '
cmd += '--overflow-list-size=20000000 '
cmd += '--io-buffer-size=256 {} {} 2> {}'
cmd = cmd.format(nth, bam, dupmark_bam, dup_qc)
run_shell_cmd(cmd)
return dupmark_bam, dup_qc
def rm_dup_pe(dupmark_bam, nth, out_dir):
prefix = os.path.join(out_dir,
os.path.basename(strip_ext_bam(dupmark_bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'dupmark')
nodup_bam = '{}.nodup.bam'.format(prefix)
run_shell_cmd(
'samtools view -F 1804 -f 2 -b {dupmark_bam} {res_param} > {nodup_bam}'
.format(
dupmark_bam=dupmark_bam,
res_param=get_samtools_res_param('view', nth=nth),
nodup_bam=nodup_bam,
))
return nodup_bam
def pbc_qc_pe(bam, mito_chr_name, nth, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
pbc_qc = '{}.lib_complexity.qc'.format(prefix)
nmsrt_bam = samtools_name_sort(bam, nth, out_dir)
cmd3 = 'bedtools bamtobed -bedpe -i {} | '
cmd3 += 'awk \'BEGIN{{OFS="\\t"}}{{print $1,$2,$4,$6,$9,$10}}\' | '
cmd3 += 'grep -v "^{}\\s" | sort | uniq -c | '
cmd3 += 'awk \'BEGIN{{mt=0;m0=0;m1=0;m2=0}} ($1==1){{m1=m1+1}} '
cmd3 += '($1==2){{m2=m2+1}} {{m0=m0+1}} {{mt=mt+$1}} END{{m1_m2=-1.0; '
cmd3 += 'if(m2>0) m1_m2=m1/m2; m0_mt=0; '
cmd3 += 'if (mt>0) m0_mt=m0/mt; m1_m0=0; if (m0>0) m1_m0=m1/m0; '
cmd3 += 'printf "%d\\t%d\\t%d\\t%d\\t%f\\t%f\\t%f\\n"'
cmd3 += ',mt,m0,m1,m2,m0_mt,m1_m0,m1_m2}}\' > {}'
cmd3 = cmd3.format(nmsrt_bam, mito_chr_name, pbc_qc)
run_shell_cmd(cmd3)
rm_f(nmsrt_bam)
return pbc_qc
def blacklist_filter_bam(bam, blacklist, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
filtered = '{}.bfilt.bam'.format(prefix)
if blacklist == '' or get_num_lines(blacklist) == 0:
cmd = 'zcat -f {} | gzip -nc > {}'.format(bam, filtered)
run_shell_cmd(cmd)
else:
# due to bedtools bug when .gz is given for -a and -b
tmp2 = gunzip(blacklist, 'tmp2', out_dir)
cmd = 'bedtools intersect -nonamecheck -v -abam {} -b {} > {}'
cmd = cmd.format(
bam,
tmp2, # blacklist
filtered)
run_shell_cmd(cmd)
rm_f([tmp2])
return filtered
def pbc_qc_se(bam, mito_chr_name, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'dupmark')
pbc_qc = '{}.lib_complexity.qc'.format(prefix)
cmd2 = 'bedtools bamtobed -i {} | '
cmd2 += 'awk \'BEGIN{{OFS="\\t"}}{{print $1,$2,$3,$6}}\' | '
cmd2 += 'grep -v "^{}\\s" | sort | uniq -c | '
cmd2 += 'awk \'BEGIN{{mt=0;m0=0;m1=0;m2=0}} ($1==1){{m1=m1+1}} '
cmd2 += '($1==2){{m2=m2+1}} {{m0=m0+1}} '
cmd2 += '{{mt=mt+$1}} END{{m1_m2=-1.0; '
cmd2 += 'if(m2>0) m1_m2=m1/m2; m0_mt=0; '
cmd2 += 'if (mt>0) m0_mt=m0/mt; m1_m0=0; if (m0>0) m1_m0=m1/m0; '
cmd2 += 'printf "%d\\t%d\\t%d\\t%d\\t%f\\t%f\\t%f\\n",'
cmd2 += 'mt,m0,m1,m2,m0_mt,m1_m0,m1_m2}}\' > {}'
cmd2 = cmd2.format(bam, mito_chr_name, pbc_qc)
run_shell_cmd(cmd2)
return pbc_qc
def rm_unmapped_lowq_reads_se(bam, multimapping, mapq_thresh, nth, mem_gb,
out_dir):
"""There are pipes with multiple samtools commands.
For such pipes, use multiple threads (-@) for only one of them.
Priority is on sort > index > fixmate > view.
"""
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
filt_bam = '{}.filt.bam'.format(prefix)
if multimapping:
qname_sort_bam = samtools_name_sort(bam, nth, mem_gb, out_dir)
run_shell_cmd(
'samtools view -h {qname_sort_bam} | '
'$(which assign_multimappers.py) -k {multimapping} | '
'samtools view -F 1804 -Su /dev/stdin | '
'samtools sort /dev/stdin -o {filt_bam} -T {prefix} {res_param}'.
format(
qname_sort_bam=qname_sort_bam,
multimapping=multimapping,
filt_bam=filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort',
nth=nth,
mem_gb=mem_gb),
))
rm_f(qname_sort_bam) # remove temporary files
else:
run_shell_cmd(
'samtools view -F 1804 -q {mapq_thresh} -u {bam} | '
'samtools sort /dev/stdin -o {filt_bam} -T {prefix} {res_param}'.
format(
mapq_thresh=mapq_thresh,
bam=bam,
filt_bam=filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort',
nth=nth,
mem_gb=mem_gb),
))
return filt_bam
def main():
# read params
args = parse_arguments()
CHROMSIZES = args.chrsz
TSS = args.tss if args.tss and os.path.basename(args.tss) != 'null' else ''
FINAL_BAM = args.nodup_bam
OUTPUT_PREFIX = os.path.join(args.out_dir,
os.path.basename(strip_ext_bam(FINAL_BAM)))
samtools_index(FINAL_BAM) # make an index first
RG_FREE_FINAL_BAM = remove_read_group(FINAL_BAM)
log.info('Initializing and making output directory...')
mkdir_p(args.out_dir)
# Also get read length
# read_len = get_read_length(FASTQ)
if args.read_len_log:
with open(args.read_len_log, 'r') as fp:
read_len = int(fp.read().strip())
elif args.read_len:
read_len = args.read_len
else:
read_len = None
# Enrichments: V plot for enrichment
# Use final to avoid duplicates
tss_plot, tss_large_plot, tss_enrich_qc = \
make_tss_plot(FINAL_BAM,
TSS,
OUTPUT_PREFIX,
CHROMSIZES,
read_len)
# remove temporary files
rm_f(RG_FREE_FINAL_BAM)
log.info('List all files in output directory...')
ls_l(args.out_dir)
log.info('All done.')
def main():
# read params
args = parse_arguments()
FINAL_BAM = args.nodup_bam
OUTPUT_PREFIX = os.path.join(args.out_dir,
os.path.basename(strip_ext_bam(FINAL_BAM)))
RG_FREE_FINAL_BAM = remove_read_group(FINAL_BAM)
# Insert size distribution - CAN'T GET THIS FOR SE FILES
insert_data, insert_plot = get_insert_distribution(RG_FREE_FINAL_BAM,
OUTPUT_PREFIX)
# Also need to run n-nucleosome estimation
fragment_length_qc(read_picard_histogram(insert_data), OUTPUT_PREFIX)
fragment_length_plot(insert_data, OUTPUT_PREFIX)
rm_f(RG_FREE_FINAL_BAM)
log.info('List all files in output directory...')
ls_l(args.out_dir)
log.info('All done.')
def mark_dup_picard(bam, out_dir): # shared by both se and pe
prefix = os.path.join(out_dir,
os.path.basename(strip_ext_bam(bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'filt')
dupmark_bam = '{}.dupmark.bam'.format(prefix)
dup_qc = '{}.dup.qc'.format(prefix)
cmd = 'java -Xmx4G -XX:ParallelGCThreads=1 -jar '
cmd += locate_picard()
cmd += ' MarkDuplicates '
# cmd = 'picard MarkDuplicates '
cmd += 'INPUT={} OUTPUT={} '
cmd += 'METRICS_FILE={} VALIDATION_STRINGENCY=LENIENT '
cmd += 'USE_JDK_DEFLATER=TRUE USE_JDK_INFLATER=TRUE '
cmd += 'ASSUME_SORTED=true REMOVE_DUPLICATES=false'
cmd = cmd.format(
bam,
dupmark_bam,
dup_qc)
run_shell_cmd(cmd)
return dupmark_bam, dup_qc
def pbc_qc_se(bam, mito_chr_name, mem_gb, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
# strip extension appended in the previous step
prefix = strip_ext(prefix, 'dupmark')
pbc_qc = '{}.lib_complexity.qc'.format(prefix)
run_shell_cmd(
'bedtools bamtobed -i {bam} | '
'awk \'BEGIN{{OFS="\\t"}}{{print $1,$2,$3,$6}}\' | '
'grep -v "^{mito_chr_name}\\s" | sort {sort_param} | uniq -c | '
'awk \'BEGIN{{mt=0;m0=0;m1=0;m2=0}} ($1==1){{m1=m1+1}} '
'($1==2){{m2=m2+1}} {{m0=m0+1}} '
'{{mt=mt+$1}} END{{m1_m2=-1.0; '
'if(m2>0) m1_m2=m1/m2; m0_mt=0; '
'if (mt>0) m0_mt=m0/mt; m1_m0=0; if (m0>0) m1_m0=m1/m0; '
'printf "%d\\t%d\\t%d\\t%d\\t%f\\t%f\\t%f\\n",'
'mt,m0,m1,m2,m0_mt,m1_m0,m1_m2}}\' > {pbc_qc}'.format(
bam=bam,
mito_chr_name=mito_chr_name,
sort_param=get_gnu_sort_param(mem_gb * 1024**3, ratio=0.5),
pbc_qc=pbc_qc,
))
return pbc_qc
def pbc_qc_pe(bam, mito_chr_name, nth, mem_gb, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
pbc_qc = '{}.lib_complexity.qc'.format(prefix)
nmsrt_bam = samtools_name_sort(bam, nth, mem_gb, out_dir)
run_shell_cmd(
'bedtools bamtobed -bedpe -i {nmsrt_bam} | '
'awk \'BEGIN{{OFS="\\t"}}{{print $1,$2,$4,$6,$9,$10}}\' | '
'grep -v "^{mito_chr_name}\\s" | sort {sort_param} | uniq -c | '
'awk \'BEGIN{{mt=0;m0=0;m1=0;m2=0}} ($1==1){{m1=m1+1}} '
'($1==2){{m2=m2+1}} {{m0=m0+1}} {{mt=mt+$1}} END{{m1_m2=-1.0; '
'if(m2>0) m1_m2=m1/m2; m0_mt=0; '
'if (mt>0) m0_mt=m0/mt; m1_m0=0; if (m0>0) m1_m0=m1/m0; '
'printf "%d\\t%d\\t%d\\t%d\\t%f\\t%f\\t%f\\n"'
',mt,m0,m1,m2,m0_mt,m1_m0,m1_m2}}\' > {pbc_qc}'.format(
nmsrt_bam=nmsrt_bam,
mito_chr_name=mito_chr_name,
sort_param=get_gnu_sort_param(mem_gb * 1024**3, ratio=0.5),
pbc_qc=pbc_qc,
))
rm_f(nmsrt_bam)
return pbc_qc
def bam_to_pbam(bam, ref_fa, out_dir='.'):
'''Convert BAM into pBAM.
Requirements:
- Python package `ptools_bin`
- `samtools`
'''
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
pbam_tmp = '{}.sorted.p.bam'.format(prefix)
pbam = '{}.p.bam'.format(prefix)
temp_files = []
if ref_fa.endswith('.gz'):
gunzipped_ref_fa = '{}.fasta'.format(
os.path.join(out_dir, os.path.basename(strip_ext_gz(ref_fa))))
run_shell_cmd('zcat -f {ref_fa} > {gunzipped_ref_fa}'.format(
ref_fa=ref_fa,
gunzipped_ref_fa=gunzipped_ref_fa,
))
temp_files.append(gunzipped_ref_fa)
else:
gunzipped_ref_fa = ref_fa
run_shell_cmd('makepBAM_genome.sh {bam} {gunzipped_ref_fa}'.format(
bam=bam,
gunzipped_ref_fa=gunzipped_ref_fa,
))
run_shell_cmd('mv {pbam_tmp} {pbam}'.format(
pbam_tmp=pbam_tmp,
pbam=pbam,
))
rm_f(temp_files)
return pbam
def bam2ta_pe(bam, nth, out_dir):
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
ta = '{}.tagAlign.gz'.format(prefix)
# intermediate files
bedpe = '{}.bedpe.gz'.format(prefix)
nmsrt_bam = samtools_name_sort(bam, nth, out_dir)
cmd1 = 'LC_COLLATE=C bedtools bamtobed -bedpe -mate1 -i {} | '
# cmd1 += 'sort -k1,1 -k2,2n -k3,3n | '
cmd1 += 'gzip -nc > {}'
cmd1 = cmd1.format(nmsrt_bam, bedpe)
run_shell_cmd(cmd1)
rm_f(nmsrt_bam)
cmd2 = 'zcat -f {} | '
cmd2 += 'awk \'BEGIN{{OFS="\\t"}}'
cmd2 += '{{printf "%s\\t%s\\t%s\\tN\\t1000\\t%s\\n'
cmd2 += '%s\\t%s\\t%s\\tN\\t1000\\t%s\\n",'
cmd2 += '$1,$2,$3,$9,$4,$5,$6,$10}}\' | '
cmd2 += 'gzip -nc > {}'
cmd2 = cmd2.format(bedpe, ta)
run_shell_cmd(cmd2)
rm_f(bedpe)
return ta
def rm_unmapped_lowq_reads_pe(bam, multimapping, mapq_thresh, nth, mem_gb,
out_dir):
"""There are pipes with multiple samtools commands.
For such pipes, use multiple threads (-@) for only one of them.
Priority is on sort > index > fixmate > view.
"""
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
filt_bam = '{}.filt.bam'.format(prefix)
tmp_filt_bam = '{}.tmp_filt.bam'.format(prefix)
fixmate_bam = '{}.fixmate.bam'.format(prefix)
if multimapping:
run_shell_cmd(
'samtools view -F 524 -f 2 -u {bam} | '
'samtools sort -n /dev/stdin -o {tmp_filt_bam} -T {prefix} {res_param} '
.format(
bam=bam,
tmp_filt_bam=tmp_filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('view', nth=nth),
))
run_shell_cmd(
'samtools view -h {tmp_filt_bam} | '
'$(which assign_multimappers.py) -k {multimapping} --paired-end | '
'samtools fixmate -r /dev/stdin {fixmate_bam} {res_param}'.format(
tmp_filt_bam=tmp_filt_bam,
multimapping=multimapping,
fixmate_bam=fixmate_bam,
res_param=get_samtools_res_param('fixmate', nth=nth),
))
else:
run_shell_cmd(
'samtools view -F 1804 -f 2 -q {mapq_thresh} -u {bam} | '
'samtools sort -n /dev/stdin -o {tmp_filt_bam} -T {prefix} {res_param}'
.format(
mapq_thresh=mapq_thresh,
bam=bam,
tmp_filt_bam=tmp_filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort',
nth=nth,
mem_gb=mem_gb),
))
run_shell_cmd(
'samtools fixmate -r {tmp_filt_bam} {fixmate_bam} {res_param}'.
format(
tmp_filt_bam=tmp_filt_bam,
fixmate_bam=fixmate_bam,
res_param=get_samtools_res_param('fixmate', nth=nth),
))
rm_f(tmp_filt_bam)
run_shell_cmd(
'samtools view -F 1804 -f 2 -u {fixmate_bam} | '
'samtools sort /dev/stdin -o {filt_bam} -T {prefix} {res_param}'.
format(
fixmate_bam=fixmate_bam,
filt_bam=filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort', nth=nth, mem_gb=mem_gb),
))
rm_f(fixmate_bam)
log.info('Checking if filtered (but not deduped) BAM is empty '
'after filtering with "samtools view -F 1804 -f 2".')
if bam_is_empty(filt_bam, nth):
raise ValueError(
'No reads found aftering filtering "samtools fixmate"d PE BAM with '
'"samtools view -F 1804 -f 2". '
'Reads are not properly paired even though mapping rate is good? ')
return filt_bam
def rm_unmapped_lowq_reads_pe(bam, multimapping, mapq_thresh, nth, mem_gb,
out_dir):
"""There are pipes with multiple samtools commands.
For such pipes, use multiple threads (-@) for only one of them.
Priority is on sort > index > fixmate > view.
"""
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
filt_bam = '{}.filt.bam'.format(prefix)
tmp_filt_bam = '{}.tmp_filt.bam'.format(prefix)
fixmate_bam = '{}.fixmate.bam'.format(prefix)
if multimapping:
run_shell_cmd(
'samtools view -F 524 -f 2 -u {bam} | '
'samtools sort -n /dev/stdin -o {tmp_filt_bam} -T {prefix} {res_param} '
.format(
bam=bam,
tmp_filt_bam=tmp_filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('view', nth=nth),
))
run_shell_cmd(
'samtools view -h {tmp_filt_bam} | '
'$(which assign_multimappers.py) -k {multimapping} --paired-end | '
'samtools fixmate -r /dev/stdin {fixmate_bam} {res_param}'.format(
tmp_filt_bam=tmp_filt_bam,
multimapping=multimapping,
fixmate_bam=fixmate_bam,
res_param=get_samtools_res_param('fixmate', nth=nth),
))
else:
run_shell_cmd(
'samtools view -F 1804 -f 2 -q {mapq_thresh} -u {bam} | '
'samtools sort -n /dev/stdin -o {tmp_filt_bam} -T {prefix} {res_param}'
.format(
mapq_thresh=mapq_thresh,
bam=bam,
tmp_filt_bam=tmp_filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort',
nth=nth,
mem_gb=mem_gb),
))
run_shell_cmd(
'samtools fixmate -r {tmp_filt_bam} {fixmate_bam} {res_param}'.
format(
tmp_filt_bam=tmp_filt_bam,
fixmate_bam=fixmate_bam,
res_param=get_samtools_res_param('fixmate', nth=nth),
))
rm_f(tmp_filt_bam)
run_shell_cmd(
'samtools view -F 1804 -f 2 -u {fixmate_bam} | '
'samtools sort /dev/stdin -o {filt_bam} -T {prefix} {res_param}'.
format(
fixmate_bam=fixmate_bam,
filt_bam=filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort', nth=nth, mem_gb=mem_gb),
))
rm_f(fixmate_bam)
return filt_bam
