def replace_genome(filename, tp, outfile1, outfile2, BWA):
f, out1, out2 = shell.IsGzipFile(filename), open(outfile1, 'wt'), open(
outfile2, 'wt')
for line in f:
if line[0] == '>':
out1.write(line)
out2.write(line)
else:
out1.write(line.upper().replace(tp[0], tp[1]))
out2.write(line.upper().replace(BT[tp[0]], BT[tp[1]]))
for fd in [f, out1, out2]:
fd.close()
# index and rm
p1 = subprocess.Popen([BWA, 'index', outfile1],
stderr=subprocess.PIPE,
stdout=subprocess.PIPE)
p2 = subprocess.Popen([BWA, 'index', outfile2],
stderr=subprocess.PIPE,
stdout=subprocess.PIPE)
p1.wait()
p2.wait()
if p1.returncode != 0 or p2.returncode != 0:
shell.eprint('[' + PROGRAM + '] Error: bwa index failed')
sys.exit(1)
os.system('rm -f ' + outfile1)
os.system('rm -f ' + outfile2)
return 0
def bwaaln(tp, genomefile, fq1, fq2, phred, outdir, config):
if phred == 33:
args = []
elif phred == 64:
args = ['-I']
else:
shell.eprint('[' + PROGRAM + '] Error: phred error')
sys.exit(1)
for k, v in config['bwa']['aln'][tp].items():
if v != '':
args += [k, v]
p1 = subprocess.Popen(
[BWA, 'aln'] + args +
[genomefile, fq1, '-f', outdir + os.path.basename(fq1) + '_1.sai'],
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL)
if fq2 != '':
p2 = subprocess.Popen(
[BWA, 'aln'] + args +
[genomefile, fq2, '-f', outdir + os.path.basename(fq2) + '_2.sai'],
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL)
p2.wait()
p1.wait()
if p1.returncode != 0 and p2.returncode != 0:
shell.eprint('[' + PROGRAM + '] Error: bwa aln error')
sys.exit(1)
args = []
for k, v in config['bwa']['sampe'][tp].items():
args += [k, v]
if fq2 != '':
p = subprocess.Popen([BWA, 'sampe'] + args + [
genomefile, outdir + os.path.basename(fq1) + '_1.sai',
outdir + os.path.basename(fq2) + '_2.sai', fq1, fq2
],
stdout=subprocess.PIPE,
stderr=subprocess.DEVNULL)
else:
p = subprocess.Popen([
BWA, 'samse', genomefile,
outdir + os.path.basename(fq1) + '_1.sai', fq1
],
stdout=subprocess.PIPE,
stderr=subprocess.DEVNULL)
infile = pysam.AlignmentFile(p.stdout, 'r')
oufile = pysam.AlignmentFile(outdir + tp + '_aln.bam',
'wb',
template=infile)
for r in infile:
oufile.write(r)
p.wait()
oufile.close()
infile.close()
os.system('rm -f ' + outdir + os.path.basename(fq1) + '_1.sai')
if fq2:
os.system('rm -f ' + outdir + os.path.basename(fq2) + '_2.sai')
return outdir + tp + '_aln.bam'
def spawnSourceFiles(entries, isTarget=False, systemHeaders=False):
if not entries:
return
for path in entries:
if os.path.isfile(path):
spawnSourceFile(path, isTarget, systemHeaders)
elif os.path.isdir(path):
spawnSourceDirectory(path, isTarget, systemHeaders)
else:
eprint("Error: couldn't process path '%s'." % path)
def _transfer_cigar(cigartuples, start, query, qual):
'''
Judge the extend end of scaffold by read if out, and return
'''
#shell.eprint(cigartuples)
d = {} # record handle of each cigar
offset = 0 # the offset on read
i = start # i for Delete position in genome
#shell.eprint(query)
for operation, num in cigartuples:
#shell.eprint(offset)
#shell.eprint(query[offset: offset+num])
if operation == 0: ## M
try:
d['M'].append(
(start, start + num, query[offset:offset + num],
qual[offset:offset + num]))
except KeyError:
d['M'] = [(start, start + num, query[offset:offset + num],
qual[offset:offset + num])]
#shell.eprint(d)
start += num - 1
i += num
offset += num
elif operation == 1: ## I
try:
d['I'].append((start, query[offset:offset + num],
qual[offset:offset + num]))
except KeyError:
d['I'] = [(start, query[offset:offset + num],
qual[offset:offset + num])]
start += 1
offset += num
elif operation == 2: ## D
try:
d['D'].append((i, num))
except KeyError:
d['D'] = [(i, num)]
start += num + 1
i += num
elif operation == 3: ## N
start += num + 1
i += num
offset += num
elif operation == 4: ## S
offset += num
else:
shell.eprint(
'[' + PROGRAM +
'] Error: the cigar string may have unsupported char, should only have [MISDN]'
)
return start, d
def bamsortindex(bamfile):
prefix = '.'.join(bamfile.split('.')[:-1])
try:
pysam.sort('-o', prefix + '.sorted.bam', bamfile)
except:
shell.eprint('[' + PROGRAM + '] Error: ' + bamfile + ' sort error')
sys.exit(1)
try:
pysam.index(prefix + '.sorted.bam')
except:
shell.eprint('[' + PROGRAM + '] Error: ' + bamfile + ' index error')
os.system('rm -f ' + bamfile)
return prefix + '.sorted.bam'
def _trim(cigartuples, t, start, seq, qual):
'''
Trim the reads' 5' and 3' end.
'''
newcigarlist, flag, tr = [], 1, 0 ## tr
for operation, num in cigartuples:
if flag:
if operation in (0, 1, 4): ## 0 for M, 1 for I, 4 for S
if num + tr > t[0]:
#shell.eprint(num, tr, t[0])
newcigarlist.append((operation, num + tr - t[0]))
if operation == 0 and len(t) == 2:
start += t[0] - tr
flag = 0
else:
tr += num
if operation == 0 and len(t) == 2:
start += num
#shell.eprint(newcigarlist, 'ok')
elif operation in (2, 3): ## 2 for D, 3 for N
if len(t) == 2:
start += num
else:
shell.eprint(
'[' + PROGRAM +
'] Error: the cigar string may have unsupported char, should only have [MISDN]'
)
else:
newcigarlist.append((operation, num))
#shell.eprint(newcigarlist, t)
if len(t) == 2:
return _trim(newcigarlist[::-1], [t[1]], start, seq[t[0]:-t[1]],
qual[t[0]:-t[1]])
elif len(t) == 1:
##
if newcigarlist[0][0] == 1:
newcigarlist[0] = (4, newcigarlist[0][1])
elif newcigarlist[0][0] in (2, 3):
newcigarlist = newcigarlist[1:]
##
if newcigarlist[-1][0] == 1:
newcigarlist[-1] = (4, newcigarlist[-1][1])
elif newcigarlist[-1][0] in (2, 3):
newcigarlist = newcigarlist[:-1]
return newcigarlist[::-1], start, seq, qual
def getDotPicture(graph, engine):
if not executableExists(engine):
eprint("No 'dot' executable!")
dotFileName = None
pngFileName = None
with tempfile.NamedTemporaryFile(delete=False) as dotFile:
dotFileName = os.path.abspath(dotFile.name)
dotFile.write(graph)
with tempfile.NamedTemporaryFile(delete=False) as pngFile:
pngFileName = os.path.abspath(pngFile.name)
runCommand("%s -Tpng -o %s %s" % (engine, pngFileName, dotFileName))
with open(pngFileName, "r") as graphFile:
return graphFile.read()
def get_config():
try:
optlist, args = getopt.getopt(sys.argv[1:], 'hp:t:',
['help', 'trim=', 'prefix='])
except getopt.GetoptError as e:
shell.eprint('[' + PROGRAM + '] Error: ' + str(e))
sys.exit(2)
if optlist == [] and args == []:
print_help()
sys.exit(0)
config = {'sam2base': {}}
for opt, value in optlist:
if opt in ('-h', '--help'):
print_help()
sys.exit(0)
elif opt in ('-t', '--trim'):
try:
config['sam2base']['trim'] = (int(value.split(',')[0]),
int(value.split(',')[1]))
except ValueError:
shell.eprint('[' + PROGRAM +
'] Error: --trim parameter should be integer')
sys.exit(1)
elif opt in ('p', '--prefix'):
if value.endswith('.'):
shell.eprint(
'[' + PROGRAM +
'] Error: --prefix parameter should not end with \'.\'')
sys.exit(1)
else:
config['sam2base']['output'] = value
else:
assert False, 'unhandled option'
try:
genomefile, bamfile = args
except ValueError as e:
shell.eprint('[' + PROGRAM +
'] Error: only two input files should be provided')
sys.exit(1)
return config, genomefile, bamfile
def soapnuke(tp, fq1, fq2, phred, config):
if phred == 33:
args = ['-Q', '2']
elif phred == 64:
args = ['-Q', '1']
else:
shell.eprint('[' + PROGRAM + '] Error: phred error')
sys.exit(1)
if fq2 == '':
args += [
'-1', fq1, '-o', 'soapnuke/' + tp, '-C',
os.path.basename(fq1) + '.clean.fq.gz'
]
fq1 = 'soapnuke/' + tp + '/' + os.path.basename(fq1) + '.clean.fq.gz'
else:
args += [
'-1', fq1, '-2', fq2, '-o', 'soapnuke/' + tp, '-C',
os.path.basename(fq1) + '_1.clean.fq.gz', '-D',
os.path.basename(fq2) + '_2.clean.fq.gz'
]
fq1, fq2 = 'soapnuke/' + tp + '/' + os.path.basename(
fq1) + '_1.clean.fq.gz', 'soapnuke/' + tp + '/' + os.path.basename(
fq2) + '_2.clean.fq.gz'
for k, v in config['soapnuke']['filter'][tp].items():
if v != '':
args += [k, v]
p = subprocess.Popen([SOAPNUKE, 'filter'] + args + ['-G', '-5', '1'],
stdout=open('log', 'w'),
stderr=subprocess.STDOUT)
p.wait()
if p.returncode != 0:
shell.eprint('[' + PROGRAM + '] Error: soapnuke run error')
shell.eprint(''.join(open('log', 'r').read()))
sys.exit(1)
return fq1, fq2, 33
def sam2base(genomefile, sortbamfile, trim=(0, 0), output='', suffix='.sb.gz'):
'''
Bam to sb file. sb -> singlebase, for the future, sbz for the sb specific compressed file,
sbi for the sb index file.
'''
def _trim(cigartuples, t, start, seq, qual):
'''
Trim the reads' 5' and 3' end.
'''
newcigarlist, flag, tr = [], 1, 0 ## tr
for operation, num in cigartuples:
if flag:
if operation in (0, 1, 4): ## 0 for M, 1 for I, 4 for S
if num + tr > t[0]:
#shell.eprint(num, tr, t[0])
newcigarlist.append((operation, num + tr - t[0]))
if operation == 0 and len(t) == 2:
start += t[0] - tr
flag = 0
else:
tr += num
if operation == 0 and len(t) == 2:
start += num
#shell.eprint(newcigarlist, 'ok')
elif operation in (2, 3): ## 2 for D, 3 for N
if len(t) == 2:
start += num
else:
shell.eprint(
'[' + PROGRAM +
'] Error: the cigar string may have unsupported char, should only have [MISDN]'
)
else:
newcigarlist.append((operation, num))
#shell.eprint(newcigarlist, t)
if len(t) == 2:
return _trim(newcigarlist[::-1], [t[1]], start, seq[t[0]:-t[1]],
qual[t[0]:-t[1]])
elif len(t) == 1:
##
if newcigarlist[0][0] == 1:
newcigarlist[0] = (4, newcigarlist[0][1])
elif newcigarlist[0][0] in (2, 3):
newcigarlist = newcigarlist[1:]
##
if newcigarlist[-1][0] == 1:
newcigarlist[-1] = (4, newcigarlist[-1][1])
elif newcigarlist[-1][0] in (2, 3):
newcigarlist = newcigarlist[:-1]
return newcigarlist[::-1], start, seq, qual
def _transfer_cigar(cigartuples, start, query, qual):
'''
Judge the extend end of scaffold by read if out, and return
'''
#shell.eprint(cigartuples)
d = {} # record handle of each cigar
offset = 0 # the offset on read
i = start # i for Delete position in genome
#shell.eprint(query)
for operation, num in cigartuples:
#shell.eprint(offset)
#shell.eprint(query[offset: offset+num])
if operation == 0: ## M
try:
d['M'].append(
(start, start + num, query[offset:offset + num],
qual[offset:offset + num]))
except KeyError:
d['M'] = [(start, start + num, query[offset:offset + num],
qual[offset:offset + num])]
#shell.eprint(d)
start += num - 1
i += num
offset += num
elif operation == 1: ## I
try:
d['I'].append((start, query[offset:offset + num],
qual[offset:offset + num]))
except KeyError:
d['I'] = [(start, query[offset:offset + num],
qual[offset:offset + num])]
start += 1
offset += num
elif operation == 2: ## D
try:
d['D'].append((i, num))
except KeyError:
d['D'] = [(i, num)]
start += num + 1
i += num
elif operation == 3: ## N
start += num + 1
i += num
offset += num
elif operation == 4: ## S
offset += num
else:
shell.eprint(
'[' + PROGRAM +
'] Error: the cigar string may have unsupported char, should only have [MISDN]'
)
return start, d
def _update(seq, dw, **kw):
if 'M' in kw:
for line in kw['M']:
#shell.eprint(line)
for j, i in enumerate(range(line[0], line[1])):
if i not in dw:
dw[i] = {'M': [line[2][j], line[3][j]]}
else:
try:
dw[i]['M'][0] += line[2][j]
dw[i]['M'][1] += line[3][j]
except KeyError:
#shell.eprint(j)
dw[i].update({'M': [line[2][j], line[3][j]]})
if 'I' in kw:
for line in kw['I']:
if line[0] not in dw:
dw[line[0]] = {'I': [line[1], line[2]]}
else:
try:
dw[line[0]]['I'][0] += ',' + line[1]
dw[line[0]]['I'][1] += ',' + line[2]
except KeyError:
dw[line[0]].update({'I': [line[1], line[2]]})
if 'D' in kw:
for line in kw['D']:
st = ''
for i in range(line[1]):
st += seq[line[0] + i - 1]
if line[0] not in dw:
dw[line[0]] = {'D': [st, '0' * line[1]]}
else:
try:
dw[line[0]]['D'][0] += ',' + st
dw[line[0]]['D'][1] += ',' + '0' * line[1]
except KeyError:
dw[line[0]].update({'D': [st, '0' * line[1]]})
return dw
def _write2gzipfile(dw, fw):
wr = []
for key in sorted(dw.keys()):
if 'M' in dw[key]:
if 'I' in dw[key]:
if 'D' in dw[key]:
wr.append(sca + '\t' + str(key) + '\t' + seq[key - 1] +
'\t' + '[M]:' + dw[key]['M'][0] + ';[I]:' +
dw[key]['I'][0] + ';[D]:' + dw[key]['D'][0] +
'\t' + '[M]:' + dw[key]['M'][1] + ';[I]:' +
dw[key]['I'][1] + ';[D]:' + dw[key]['D'][1] +
'\t' + str(len(dw[key]['M'][0])) + '\n')
else:
wr.append(sca + '\t' + str(key) + '\t' + seq[key - 1] +
'\t' + '[M]:' + dw[key]['M'][0] + ';[I]:' +
dw[key]['I'][0] + '\t' + '[M]:' +
dw[key]['M'][1] + ';[I]:' + dw[key]['I'][1] +
'\t' + str(len(dw[key]['M'][0])) + '\n')
elif 'D' in dw[key]:
wr.append(sca + '\t' + str(key) + '\t' + seq[key - 1] +
'\t' + '[M]:' + dw[key]['M'][0] + ';[D]:' +
dw[key]['D'][0] + '\t' + '[M]:' +
dw[key]['M'][1] + ';[D]:' + dw[key]['D'][1] +
'\t' + str(len(dw[key]['M'][0])) + '\n')
else:
wr.append(sca + '\t' + str(key) + '\t' + seq[key - 1] +
'\t' + '[M]:' + dw[key]['M'][0] + '\t' + '[M]:' +
dw[key]['M'][1] + '\t' +
str(len(dw[key]['M'][0])) + '\n')
fw.write(''.join(wr))
return {}
if not sortbamfile.endswith('.bam'):
shell.eprint('[' + PROGRAM +
'] Error: input should be sorted and indexed bam')
sys.exit(1)
if trim[0] < 0 or trim[1] < 0:
shell.eprint(
'Are you serious ? --trim should not be negative integer.')
sys.exit(255)
if output == '':
output = sortbamfile
# Get a chromosome or scaffold from genome then do with bam
## If scaffold has no mapping reads, could it raise error?
fw = gzip.open(output + suffix, 'wt')
with pysam.AlignmentFile(sortbamfile, 'rb') as f:
for sca, seq, seqlen in shell.Fa2Geno(genomefile):
dw, newend = {}, 1
try:
for read in f.fetch(contig=sca):
if 1 < newend < read.reference_start + 1:
dw = _write2gzipfile(dw, fw)
if len(read.query_sequence) != len(read.qual):
continue
#shell.eprint(read.query_sequence)
if trim == (0, 0):
newstart, d = _transfer_cigar(read.cigartuples,
read.reference_start + 1,
read.query_sequence,
read.qual)
else:
if read.flag & 16:
trim = trim[::-1]
newstart, d = _transfer_cigar(
*_trim(list(
read.cigartuples), trim, read.reference_start +
1, read.query_sequence, read.qual))
#
if seqlen < newstart:
continue
if newstart > newend:
newend = newstart
dw = _update(seq, dw, **d)
_write2gzipfile(dw, fw)
except ValueError as e:
shell.eprint('[' + PROGRAM +
'] Error: input should be sorted and indexed bam')
os.system('rm -f ' + output + suffix)
sys.exit(1)
fw.close()
return output + suffix
def merge_RES(Config,
Genomefile,
Phred_DNA=33,
Phred_RNA=33,
Qual_cutoff=30,
HomoPrior=0.99,
Rate=2,
Method='Bayesian',
Ploidy=2,
Intron=None,
DNAdepth=10,
RNAdepth=3,
Bayesian_Posterior_Probability=0.95,
FDR_DNA_Heterozygosis=0.05,
Non_Ref_BaseCount=0,
Paralogous_D=1,
Intronic=6,
Homopolymer=1,
out_path='./'):
'''
merge the result of scanner and check
'''
BT = {'A': 'T', 'T': 'A', 'C': 'G', 'G': 'C', 'N': 'N'}
BASE = ('A', 'C', 'G', 'T')
def read(f, s, table, several_tp):
for line in f:
if line[0] == '#':
continue
fd = line.strip().split()
key1, key2, key3, key4, key5 = (fd[0], fd[1], fd[2]), (s, 'DNA'), (
s, 'RNA'), (s, 'SNPvalue'), (s, 'Editvalue')
try:
table[key1].update({key2: fd[5]})
except KeyError:
table[key1] = {key2: fd[5]}
table[key1][key3] = fd[9]
if Method == 'Bayesian':
table[key1][key4] = (fd[6], fd[7])
else:
table[key1][key4] = fd[6]
table[key1][key5] = fd[12]
table[key1]['gbase'] = fd[3].upper()
try:
if table[key1]['type'] != fd[10]:
several_tp.add(key1)
except KeyError:
table[key1]['type'] = fd[10]
return 0
def deleteintron(sca, l, table):
for i in l:
try:
del table[(sca, str(i), '+')]
except KeyError:
pass
try:
del table[(sca, str(i), '-')]
except KeyError:
pass
try:
del table[(sca, str(i), '.')]
except KeyError:
pass
return 0
def caculus(f, table, errorRate, st):
for line in f:
fd = line.strip().split()
ref = fd[2]
for strand in st:
key1 = (fd[0], fd[1], strand)
if key1 in table:
key2 = (s, 'RNA')
if key2 not in table[key1]:
editV = (s, 'Editvalue')
if fd[5] == 0:
table[key1][key2] = '0,0,0,0'
table[key1][editV] = '1'
else:
seq, qual = shell._DealBasequality(
fd[3][4:4 + int(fd[-1])].upper(),
fd[4][4:4 + int(fd[-1])], Phred_RNA)
l = [seq.count(i) for b in BASE]
table[key1][key2] = '{},{},{},{}'.format(*l)
tol, editdep = 0, 0
for i in range(4):
tol += l[i]
if BASE[i] != ref and l[i] > editdep:
editdep = l[i]
table[key1][editV] = shell._SNVvalue_Binomial(
(editdep, tol), errorRate, RNA=True)
return 0
if Method not in ('Bayesian', 'Binomial', 'Frequency'):
print_help()
sys.exit(1)
elif Method == 'Bayesian' and Genomefile == None:
print_help()
sys.exit(1)
HetePrior = 1 - HomoPrior
## read config file
sample, order = {}, []
with open(Config, 'r') as f:
for line in f:
fd = line.strip().split()
order.append(fd[0])
if os.path.isfile(fd[1] + '.stat') == True:
sample[fd[0]] = {3: fd[1] + '.stat'}
else:
shell.eprint(
'[bigtable] Error: DNA single base file do not have .stat file in the same directory'
)
sys.exit(1)
sample[fd[0]][0] = fd[1]
if len(fd) < 4:
shell.eprint('[bigtable] Error: config file format error')
sys.exit(1)
sample[fd[0]][1.1] = fd[2]
sample[fd[0]][1.2] = fd[3]
if len(fd) == 6:
sample[fd[0]][2.1] = fd[4]
sample[fd[0]][2.2] = fd[5]
## read table
several_tp, table, peak_dep = set(), {}, {}
for s, v in sample.items():
f = shell.IsGzipFile(v[1.2])
f.readline()
read(f, s, table, several_tp)
f.close()
with open(v[3], 'r') as f:
fd = f.readlines()[1].strip().split()
peak_dep[s] = float(
fd[4]) if float(fd[4]) > float(fd[3]) else float(fd[3])
try:
f = shell.IsGzipFile(v[2.2])
f.readline()
read(f, s, table, several_tp)
f.close()
except:
pass
## delete sites have different editing type in different sample
for key in several_tp:
del table[key]
## filter sites locating near junctions
if Intron != None:
f = shell.IsGzipFile(Intron)
for line in f:
fd = line.strip().split()
if int(fd[3]) < int(fd[4]):
beg, end = int(fd[3]), int(fd[4])
else:
beg, end = int(fd[4]), int(fd[3])
deleteintron(fd[1], range(beg, beg + Intronic), table)
deleteintron(fd[1], range(end - Intronic + 1, end + 1), table)
f.close()
## filter homopolymer
if Homopolymer:
seq, leng, st = {}, {}, ''
m = [re.compile(i * 5) for i in BASE]
f = shell.IsGzipFile(Genomefile)
for line in f:
if line[0] == '>':
if st:
seq[key] = st
leng[key] = len(st)
st = ''
key = line.strip().split()[0][1:]
else:
st += line.strip().upper()
if st:
seq[key] = st
leng[key] = len(st)
delkey = set()
for key in table:
sca, pos, strand = key
beg = int(pos) - 4 if int(pos) - 4 > 1 else 1
end = int(pos) + 4 if int(pos) + 4 < leng[sca] else leng[sca]
nt = seq[sca][beg - 1:end]
for m1 in m:
if m1.match(nt):
delkey.add(key)
break
for key in delkey:
del table[key]
del seq
del leng
del delkey
if Method == 'Bayesian':
basecontent = shell._BasePercent(Genomefile)
weight = 0.5
weight_other = (1 - weight) / 2
## adjust substitution rate for illumina
FixedError = {
'A': {
'C': weight,
'T': weight_other,
'G': weight_other
},
'C': {
'A': weight,
'T': weight_other,
'G': weight_other
},
'G': {
'T': weight,
'A': weight_other,
'C': weight_other
},
'T': {
'G': weight,
'A': weight_other,
'C': weight_other
}
}
FixedKey = []
for key in FixedError.keys():
for ke in FixedError[key].keys():
if (key, ke) not in FixedKey and (ke, key) not in FixedKey:
FixedKey.append((key, ke))
errorRate = 10**(-1 * Qual_cutoff / 10)
for s, v in sample.items():
with gzip.open(v[0], 'rt') as f:
for line in f:
fd = line.strip().split()
ref = fd[2].upper()
for strand in ('+', '-', '.'):
key1 = (fd[0], fd[1], strand)
if key1 in table:
key2 = (s, 'DNA')
if key2 not in table[key1]:
snpV = (s, 'SNPvalue')
if fd[5] == 0:
table[key1][snpV] = 'NA'
table[key1][key2] = '0,0,0,0'
else:
seq, qual = shell._DealBasequality(
fd[3][4:4 + int(fd[-1])].upper(),
fd[4][4:4 + int(fd[-1])], Phred_DNA)
table[key1][key2] = ','.join(
[str(seq.count(i)) for b in BASE])
if seq == '':
table[key1][snpV] = 'NA'
continue
if Method == 'Bayesian':
result = shell._SNPvalue_Bayesian(
seq, qual, Ploidy, FixedKey,
FixedError, basecontent)
table[key1][snpV] = (result[0], result[1])
elif Method == 'Binomial':
table[key1][
snpV] = shell._SNVvalue_Binomial(
','.join([
str(seq.count(i)) for b in BASE
]), ref)
elif Method == 'Frequency':
table[key1][
snpV] = shell._SNPvalue_Frequency(
seq, ref)[1]
else:
shell.eprint(
'[bigtable] Error: --method not recognize'
)
sys.exit(1)
with gzip.open(v[1.1], 'rt') as f:
caculus(f, table, errorRate, ['+', '.'])
if 2.1 not in v.keys():
continue
with gzip.open(v[2.1], 'rt') as f:
caculus(f, table, errorRate, ['-'])
## Complement DNA and RNA information for all sites in the table
for k, v in table.items():
for spl in order:
keyDNA, keysnpV, keyRNA, keyeditV = (spl, 'DNA'), (
spl, 'SNPvalue'), (spl, 'RNA'), (spl, 'Editvalue')
if keyDNA not in v:
table[k][keyDNA] = '0,0,0,0'
table[k][keysnpV] = 'NA'
if keyRNA not in v:
table[k][keyRNA] = '0,0,0,0'
table[k][keyeditV] = '1'
## Remove sites with high DNA depth and multiple RNA editing types
delkey = set()
for key1 in table:
for s in order:
fd = [int(i) for i in table[key1][(s, 'DNA')].split(',')]
if Paralogous_D:
dep = sum(fd)
if dep > 2 * peak_dep[s]:
delkey.add(key1)
break
fd = [int(i) for i in table[key1][(s, 'RNA')].split(',')]
rna_count = {}
for i in range(4):
if BASE[i] != table[key1]['gbase']:
rna_count[BASE[i]] = fd[i]
key3 = sorted(rna_count.keys(),
key=lambda x: rna_count[x],
reverse=True)
if len(key3) != 3:
shell.eprint('[bigtable] Error: rna depth error')
sys.exit(1)
if rna_count[key3[0]] > 0 and (rna_count[key3[1]] /
rna_count[key3[0]]) > 0.01:
delkey.add(key1)
break
for key in delkey:
del table[key]
delkey = set()
for key1 in table:
kplus, kminus = (k[0], k[1], '+'), (k[0], k[1], '-')
if kplus in table and kminus in table:
plusdep, miusdep = 0, 0
for s in order:
plusdep += sum(
[int(i) for i in table[kplus][(s, 'RNA')].split(',')])
miusdep += sum(
[int(i) for i in table[kminus][(s, 'RNA')].split(',')])
if plusdep > miusdep:
delkey.add(kminus)
elif plusdep < miusdep:
delkey.add(kplus)
for key in delkey:
del table[key]
del delkey
fdr, binomial = {}, {}
for key1 in table:
for s in order:
try:
fdr[s].append([key1, table[key1][(s, 'Editvalue')]])
except KeyError:
fdr[s] = [[key1, table[key1][(s, 'Editvalue')]]]
if Method == 'Binomial':
try:
binomial[s].append([key1, table[key1][(s, 'SNPvalue')]])
except KeyError:
binomial[s] = [[key1, table[key1][(s, 'SNPvalue')]]]
for s, v in fdr.items():
fd = sorted(v, key=lambda x: float(x[1]), reverse=True)
fd_fdr = shell._Fdr([float(i[1]) for i in fd])
for i in range(len(fd_fdr)):
table[fd[i][0]][(s, 'Editvalue')] = fd_fdr[i]
least_dep = 1
if Method == 'Binomial':
for s, v in binomial.items():
fd = sorted(v, key=lambda x: float(x[1]), reverse=True)
fd_fdr = shell._Fdr([float(i[1]) for i in fd])
for i in range(len(fd_fdr)):
table[fd[i][0]][(s, 'SNPvalue')] = shell._FormatP(fd_fdr[i])
p = sorted(peak_dep.keys(), key=lambda x: peak_dep[x])
ratio = 1 / Ploidy
while least_dep < p[0]:
if shell._SNVvalue_Binomial(
(0, least_dep), ratio, RNA=True) < 0.05:
break
least_dep += 1
fw = open(out_path + '/RES.txt', 'w')
title = '#1.Chromosome\t2.Coordinate\t3.Strand\t4.Gbase\t5.EditType'
for idx, s in enumerate(order):
title += '\t' + str(
6 + idx * 2) + '.' + s + '.DNA_baseCount[A,C,G,T]\t' + str(
7 + idx * 2) + '.' + s + '.RNA_basecount[A,C,G,T];P_value'
title += '\n'
fw.write(title)
for key1 in sorted(table.keys(), key=lambda x: (x[0], int(x[1]))):
info = []
flag = 1
for s in order:
rnainfo = table[key1][(s, 'RNA')].split(',')
rna_dep = sum([int(i) for i in rnainfo])
dna_dep = sum([int(i) for i in table[key1][(s, 'DNA')].split(',')])
if table[key1][(s, 'SNPvalue')] != 'NA':
if Method == 'Bayesian':
if len(
set(table[key1][(s, 'SNPvalue')][0]) -
set(table[key1]['gbase'])) != 0 or float(
table[key1][(s, 'SNPvalue')]
[1]) < Bayesian_Posterior_Probability:
flag = 0
elif Method == 'Binomial':
if dna_dep < least_dep:
n = 0
for i in table[key1][(s, 'RNA')].split(','):
if int(i) == 0: n += 1
if n < 3: flag = 0
else:
if table[key1][(s,
'SNPvalue')] < FDR_DNA_Heterozygosis:
flag = 0
elif Method == 'Frequency':
if table[key1][(s, 'SNPvalue')] < Non_Ref_BaseCount:
flag = 0
if dna_dep >= DNAdepth and rna_dep >= RNAdepth and float(
table[key1][(s, 'Editvalue')]) < 0.05:
alt_base = table[key1]['type'].split('->')[1]
if key1[2] == '-':
alt_base = BT[alt_base]
base_dep = {}
for i in range(4):
base_dep[BASE[i]] = int(rnainfo[i])
if base_dep[alt_base] > 0:
info.append(table[key1][(s, 'DNA')] + '\t' +
table[key1][(s, 'RNA')] + ';' +
str(table[key1][(s, 'Editvalue')]) + '*')
else:
info.append(table[key1][(s, 'DNA')] + '\t' +
table[key1][(s, 'RNA')] + ';' +
str(table[key1][(s, 'Editvalue')]))
else:
info.append(table[key1][(s, 'DNA')] + '\t' +
table[key1][(s, 'RNA')] + ';' +
str(table[key1][(s, 'Editvalue')]))
if flag == 1:
fw.write(('{}\t' * 5).format(
*(list(key1) + [table[key1]['gbase'], table[key1]['type']])) +
'\t'.join(info) + '\n')
return 0
def get_config():
try:
optlist, args = getopt.getopt(sys.argv[1:], 'ho:',
['help', 'outdir=', 'bwa='])
except getopt.GetoptError as e:
shell.eprint('[' + PROGRAM + '] Error: ' + str(e))
sys.exit(2)
if optlist == [] and args == []:
print_help()
sys.exit(0)
global BWA, OUTDIR
for opt, value in optlist:
if opt in ('-h', '--help'):
print_help()
sys.exit(1)
elif opt in ('-o', '--outdir'):
OUTDIR = os.path.abspath(value) + '/'
elif opt == '--bwa':
if os.path.exists(value):
BWA = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: bwa path does not exist')
sys.exit(1)
try:
genomefile = args[0]
if os.path.isfile(genomefile):
genomefile = os.path.abspath(genomefile)
else:
shell.eprint('[' + PROGRAM + '] Error: genome file does not exist')
sys.exit(1)
except ValueError as e:
shell.eprint('[' + PROGRAM + '] Error: ' + str(e))
sys.exit(1)
if BWA == '':
r = subprocess.getstatusoutput('which bwa')
if r[0] == 1:
shell.eprint('[' + PROGRAM + '] Warning: lack bwa program')
sys.exit(1)
else:
BWA = r[1]
if os.path.exists(OUTDIR) == False:
try:
os.makedirs(OUTDIR)
except:
shell.eprint('[' + PROGRAM +
'] Error: outdir could not be created, please check')
sys.exit(1)
if OUTDIR == '':
OUTDIR = os.getcwd() + '/regeo'
try:
os.makedirs(OUTDIR)
except FileExistsError:
pass
except:
shell.eprint('[' + PROGRAM +
'] Error: outdir could not be created, please check')
sys.exit(1)
os.chdir(OUTDIR)
return genomefile
def main():
global DBAM, RBAM, DFQ1, DFQ2, RFQ1, RFQ2, PDFQ1, PRFQ1
config, genomefile = get_config()
logging.info('Program Start')
if FLAG & 2:
logging.info('do soapnuke')
if DFQ1 != '':
DFQ1, DFQ2, PDFQ1 = soapnuke('DNA', DFQ1, DFQ2, PDFQ1, config)
if RFQ1 != '':
RFQ1, RFQ2, PRFQ1 = soapnuke('RNA', RFQ1, RFQ2, PRFQ1, config)
logging.info('soapnuke has done')
if FLAG & 4:
logging.info('do pilon')
logging.info('check genome index')
for suffix in ['.amb', '.ann', '.bwt', '.pac', '.sa']:
if os.path.isfile(genomefile + suffix) == False:
logging.info('genome index not found')
p = subprocess.Popen([BWA, 'index', genomefile],
stdout=open('log', 'w'),
stderr=subprocess.STDOUT)
p.wait()
if p.returncode != 0:
shell.eprint('[' + PROGRAM +
'] Error: bwa index run error')
shell.eprint(''.join(open('log', 'r').read()))
sys.exit(1)
logging.info('genome index done')
break
try:
os.makedirs('pilon/')
except FileExistsError:
pass
except:
shell.eprint('[' + PROGRAM + '] Error: mkdir pilon/ error')
sys.exit(1)
args = []
for k, v in config['bwa']['mem'].items():
args += (k, v)
args += [genomefile, DFQ1] if DFQ2 == '' else [genomefile, DFQ1, DFQ2]
p = subprocess.Popen([BWA, 'mem'] + args,
stdout=subprocess.PIPE,
stderr=open('log', 'w'))
infile = pysam.AlignmentFile(p.stdout, 'r')
oufile = pysam.AlignmentFile('pilon/mem_just.bam',
'wb',
template=infile)
for r in infile:
oufile.write(r)
infile.close()
oufile.close()
p.wait()
if p.returncode != 0:
shell.eprint('[' + PROGRAM + '] Error: bwa mem run error')
shell.eprint(''.join(open('log', 'r').read()))
sys.exit(1)
DBAM_mem = bamsortindex('pilon/mem_just.bam')
try:
os.makedirs('pilon/sub/')
except FileExistsError:
pass
except:
shell.eprint('[' + PROGRAM + '] Error: mkdir pilon/sub/ error')
sys.exit(1)
shell.splitFa(genomefile, 'pilon/sub/')
f = pysam.AlignmentFile(DBAM_mem, 'rb')
for fn in [
'pilon/sub/' + bed for bed in os.listdir('pilon/sub/')
if bed.endswith('.bed')
]:
fw = pysam.AlignmentFile('.'.join(fn.split('.')[:-1]) + '.bam',
'wb',
template=f)
with open(fn, 'r') as fg:
for line in fg:
for r in f.fetch(contig=line.strip().split()[0]):
if r.flag & 256 or r.flag & 2048:
continue
fw.write(r)
fw.close()
try:
pysam.index('.'.join(fn.split('.')[:-1]) + '.bam')
except:
shell.eprint('[' + PROGRAM + '] Error: ' +
'.'.join(fn.split('.')[:-1]) + '.bam' +
' index error')
f.close()
for fn in [
'pilon/sub/' + fa for fa in os.listdir('pilon/sub/')
if fa.endswith('.fa')
]:
maxn = int(
subprocess.getstatusoutput(
'awk \'BEGIN{x};{x+=$3};END{print x}\' ' +
'.'.join(fn.split('.')[:-1]) + '.bed')[1])
maxmem = '-Xmx8g' if maxn <= 100000000 else '-Xmx16g'
p = subprocess.Popen([
JAVA, maxmem, '-jar', PILON, '--fix', 'snps', '--genome', fn,
'--bam', '.'.join(fn.split('.')[:-1]) + '.bam', '--output',
fn + '_fix_snps'
],
stdout=open('log', 'w'),
stderr=subprocess.STDOUT)
p.wait()
if p.returncode != 0:
shell.eprint('[' + PROGRAM + '] Error: pilon run error')
shell.eprint(''.join(open('log', 'r').read()))
sys.exit(1)
os.system('cat pilon/sub/*_fix_snps.fasta > ' + genomefile +
'.fix_snps.fa')
os.system('sed -i s/_pilon// ' + genomefile + '.fix_snps.fa')
os.system('rm -rf pilon/sub/*')
genomefile = genomefile + '.fix_snps.fa'
logging.info('pilon has done')
# check and do bwa aln
logging.info('do bwa aln')
logging.info('check fixed genome index')
for suffix in ['.amb', '.ann', '.bwt', '.pac', '.sa']:
if os.path.isfile(genomefile + suffix) == False:
logging.info('fixed genome not found')
p = subprocess.Popen([BWA, 'index', genomefile],
stdout=open('log', 'w'),
stderr=subprocess.STDOUT)
p.wait()
if p.returncode != 0:
shell.eprint(
'[' + PROGRAM +
'] Error: bwa index for fixed snps genome run error')
shell.eprint(''.join(open('log', 'r').read()))
sys.exit(1)
logging.info('fixed genome index done')
break
try:
os.makedirs('aln/')
except FileExistsError:
pass
except:
shell.eprint('[' + PROGRAM + '] Error: make aln directory error')
sys.exit(1)
if DFQ1 != '' and DBAM == '':
DBAM = bwaaln('DNA', genomefile, DFQ1, DFQ2, PDFQ1, 'aln/', config)
if RFQ1 != '' and RBAM == '':
RBAM = bwaaln('RNA', genomefile, RFQ1, RFQ2, PRFQ1, 'aln/', config)
logging.info('bwa aln has done')
logging.info('do get best bam')
if DBAM:
bestuniqbam(DBAM, DNA=True, **config['bestuniqbam']['DNA'])
DBAM = bamsortindex('.'.join(os.path.basename(DBAM).split('.')[:-1]) +
'.best.bam')
logging.info('The output file of DNA alignment is ' + DBAM)
if RBAM:
bestuniqbam(RBAM, RNA=True, **config['bestuniqbam']['RNA'])
if 'SS' in config['bestuniqbam']['RNA']:
bamsortindex('.'.join(os.path.basename(RBAM).split('.')[:-1]) +
'.negative.bam')
bamsortindex('.'.join(os.path.basename(RBAM).split('.')[:-1]) +
'.positive.bam')
logging.info(
'The output files of strand-specific RNA alignment are ' +
'.'.join(os.path.basename(RBAM).split('.')[:-1]) +
'.negative.bam and ' +
'.'.join(os.path.basename(RBAM).split('.')[:-1]) +
'.positive.bam')
else:
bamsortindex('.'.join(os.path.basename(RBAM).split('.')[:-1]) +
'.best.bam')
logging.info('The output file of RNA alignment is ' +
'.'.join(os.path.basename(RBAM).split('.')[:-1]) +
'.best.bam')
os.system('rm -f log')
logging.info('All things have been done! Have a good day!')
return 0
def get_config():
shortopts = 'ho:'
longopts = [
'help', 'outdir=', 'DNA-fq1=', 'DNA-fq2=', 'RNA-fq1=', 'RNA-fq2=',
'DNA-bam=', 'RNA-bam=', 'bwa=', 'soapnuke=', 'java=', 'pilon=',
'config=', 'DNA-mapQ=', 'RNA-mapQ=', 'rmdup=', 'uniq=', 'ss=',
'DNA-I=', 'RNA-I='
]
try:
optlist, args = getopt.getopt(sys.argv[1:], shortopts, longopts)
except getopt.GetoptError as e:
shell.eprint('[' + PROGRAM + '] Error: ' + str(e))
sys.exit(2)
if optlist == [] and args == []:
print_help()
sys.exit(0)
config = {'bestuniqbam': {'DNA': {}, 'RNA': {}}}
tobool = {'T': True, 'F': False}
global DBAM, DFQ1, DFQ2, RBAM, RFQ1, RFQ2
global PDFQ1, PDFQ2, PRFQ1, PRFQ2
global BWA, SOAPNUKE, JAVA, PILON
global FLAG, OUTDIR
for opt, value in optlist:
if opt in ('-h', '--help'):
print_help()
sys.exit(0)
elif opt in ('-o', '--outdir'):
OUTDIR = os.path.abspath(value) + '/'
elif opt == '--DNA-fq1':
if os.path.exists(value):
DFQ1 = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: dna fq1 file does not exist')
sys.exit(1)
elif opt == '--DNA-fq2':
if os.path.exists(value):
DFQ2 = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: dna fq2 file does not exist')
sys.exit(1)
elif opt == '--RNA-fq1':
if os.path.exists(value):
RFQ1 = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: rna fq1 file does not exist')
sys.exit(1)
elif opt == '--RNA-fq2':
if os.path.exists(value):
RFQ2 = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: rna fq2 file does not exist')
sys.exit(1)
elif opt == '--DNA-bam':
if os.path.exists(value):
DBAM = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: dna bam file does not exist')
sys.exit(1)
elif opt == '--RNA-bam':
if os.path.exists(value):
RBAM = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: rna bam file does not exist')
sys.exit(1)
elif opt == '--bwa':
if os.path.exists(value):
BWA = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: bwa path does not exist')
sys.exit(1)
elif opt == '--soapnuke':
if os.path.exists(value):
SOAPNUKE = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: soapnuke path does not exist')
sys.exit(1)
elif opt == '--java':
if os.path.exists(value):
JAVA = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: java path does not exist')
sys.exit(1)
elif opt == '--pilon':
if os.path.exists(value):
PILON = os.path.abspath(value)
else:
shell.eprint('[' + PROGRAM +
'] Error: pilon path does not exist')
sys.exit(1)
elif opt == '--config':
if os.path.exists(value):
with open(value, 'r') as f:
config.update(json.load(f))
else:
shell.eprint('[' + PROGRAM +
'] Error: config file does not exist')
sys.exit(1)
if len(
set(['-C', '-D', '-o', '-1', '-2'])
& config['soapnuke']['filter'].keys()) > 0:
shell.eprint(
'[' + PROGRAM +
'] Error: soapnuke parameters -C -D -o -1 -2 do not needed to provided'
)
sys.exit(1)
elif opt == '--DNA-mapQ':
try:
config['bestuniqbam']['DNA']['mapQ'] = int(value)
except ValueError:
shell.eprint('[' + PROGRAM +
'] Error: --DNA-mapQ should be integer')
sys.exit(1)
elif opt == '--RNA-mapQ':
try:
config['bestuniqbam']['RNA']['mapQ'] = int(value)
except ValueError:
shell.eprint('[' + PROGRAM +
'] Error: --RNA-mapQ should be integer')
sys.exit(1)
elif opt == '--rmdup':
try:
config['bestuniqbam']['RNA']['RmDup'] = tobool[value]
except KeyError:
shell.eprint('[' + PROGRAM +
'] Error: --rmdup should be T or F')
sys.exit(1)
elif opt == '--uniq':
try:
config['bestuniqbam']['RNA']['Uniq'] = tobool[value]
except KeyError:
shell.eprint('[' + PROGRAM +
'] Error: --uniq should be T or F')
sys.exit(1)
elif opt == '--ss':
try:
config['bestuniqbam']['RNA']['SS'] = tobool[value]
except KeyError:
shell.eprint('[' + PROGRAM + '] Error: --ss should be T or F')
sys.exit(1)
else:
assert False, 'unhandled option'
if BWA == '':
shell.eprint('[' + PROGRAM + '] Error: bwa is necessary')
sys.exit(1)
if PILON != '' and JAVA == '':
shell.eprint('[' + PROGRAM + '] Error: pilon process needs java')
sys.exit(1)
if os.path.exists(OUTDIR) == False:
try:
os.makedirs(OUTDIR)
except FileExistsError:
shell.eprint(
'[' + PROGRAM +
'] Warning: outdir have existed, may have some conflict')
except:
shell.eprint('[' + PROGRAM +
'] Error: outdir could not be created, please check')
sys.exit(1)
try:
genomefile = args[0]
genomefile = os.path.abspath(genomefile)
try:
os.symlink(genomefile, OUTDIR + os.path.basename(genomefile))
except FileExistsError:
pass
except:
shell.eprint('[' + PROGRAM +
'] Error: could not create soft link of genomefile')
sys.exit(1)
genomefile = os.path.basename(genomefile)
except ValueError:
shell.eprint('[' + PROGRAM + '] Error: ' + str(e))
sys.exit(1)
if (DFQ1 != '' and DBAM != '') or (DFQ1 == '' and DBAM == ''):
shell.eprint(
'[' + PROGRAM +
'] Error: DNA fastq file or bamfile should be provided only one')
sys.exit(1)
if (RFQ1 != '' and RBAM != '') or (RFQ1 == '' and RBAM == ''):
shell.eprint(
'[' + PROGRAM +
'] Error: RNA fastq file or bamfile should be provided only one')
sys.exit(1)
if (DBAM != '' and RBAM != '') and (SOAPNUKE != '' or PILON != ''):
shell.eprint(
'[' + PROGRAM +
'] Error: there are some logical error. Offering bam file means there is no need to do soapnuke or pilon'
)
sys.exit(1)
if DFQ1 != '':
PDFQ1 = shell.checkFqQuality(DFQ1)
if RFQ1 != '':
PRFQ1 = shell.checkFqQuality(RFQ1)
if SOAPNUKE != '':
FLAG += 2
if PILON != '':
FLAG += 4
os.chdir(OUTDIR)
logging.basicConfig(level=logging.INFO,
filename=OUTDIR + PROGRAM + '.log',
filemode='w',
format='%(asctime)s : %(message)s',
datefmt='%m/%d/%Y %I:%M:%S %p')
return config, genomefile
def bestuniqbam(bamfile,
mapQ=20,
DNA=None,
RNA=None,
SS=False,
RmDup=True,
Uniq=True,
outdir='./'):
'''
Filter bam to best uniq bam.
##eg: bestuniqbam(bamfile, RNA-mapQ=20, RNA=True, SS=True, RmDup=True, Uniq=True, outdir='./scanner')
'''
def _flag_split(g):
'''
According to sam format, 115 = 64 + 32 + 16 + 2 + 1.
'''
return set([g & 2**i for i in range(0, 12)])
# check bamfile
if bamfile.split('.')[-1] == 'bam':
file_type_tag = 'rb'
elif bamfile.split('.')[-1] == 'sam':
file_type_tag = 'r'
else:
shell.eprint(
'[' + PROGRAM +
'] Error: the input bamfile/samfile should be *.bam or *.sam.')
sys.exit(1)
# parameter tickle
if (RNA is None and DNA is None) or (RNA != None and DNA != None):
shell.eprint('[' + PROGRAM + '] Error: --DNA or --RNA is needed!')
sys.exit(1)
if DNA == True and SS == True:
shell.eprint(
'[' + PROGRAM +
'] Warning: DNA do not have --ss, but it\'s ok to run this program.'
)
if os.path.exists(outdir) == False:
try:
os.makedirs(outdir)
except:
sys.exit(1)
# core
try:
f = pysam.AlignmentFile(bamfile, file_type_tag)
except:
shell.eprint('[' + PROGRAM + '] Error: check the bam file ' +
os.path.basename(bamfile) + ' please!')
sys.exit(1)
## DNA and RNA's bamfile name should be different by yourself.
if SS:
fn1 = outdir + '.'.join(
os.path.basename(bamfile).split('.')[:-1]) + '.negative.bam'
fn2 = outdir + '.'.join(
os.path.basename(bamfile).split('.')[:-1]) + '.positive.bam'
fw1 = pysam.AlignmentFile(fn1, 'wb', template=f)
fw2 = pysam.AlignmentFile(fn2, 'wb', template=f)
else:
fn3 = outdir + '.'.join(
os.path.basename(bamfile).split('.')[:-1]) + '.best.bam'
fw3 = pysam.AlignmentFile(fn3, 'wb', template=f)
for line in f.fetch(until_eof=True):
if int(line.mapping_quality) < mapQ:
continue
flag = int(line.flag)
tags = line.get_tags()
if 1024 & flag and RmDup:
continue
if RNA:
if Uniq and (('XT', 'U') not in tags or ('X0', 1) not in tags or
('X1', 0) not in tags):
continue
if 256 & flag:
continue
if flag in (67, 131, 115, 179):
continue
if 32 & flag and 16 & flag:
continue
if SS:
if 64 & flag and 32 & flag and line.template_length >= 0:
fw1.write(line)
elif 128 & flag and 16 & flag and line.template_length <= 0:
fw1.write(line)
elif 64 & flag and 16 & flag and line.template_length <= 0:
fw2.write(line)
elif 128 & flag and 32 & flag and line.template_length >= 0:
fw2.write(line)
elif 64 & flag and 16 & flag != 16 and 32 & flag != 32:
fw1.write(line)
elif 128 & flag and 16 & flag != 16 and 32 & flag != 32:
fw2.write(line)
elif flag == 16:
fw2.write(line)
elif flag == 0:
fw1.write(line)
else:
fw3.write(line)
elif DNA:
fw3.write(line)
else:
pass
if SS:
fw1.close()
fw2.close()
return fn1, fn2
else:
fw3.close()
return fn3