def main():
parser = argparse.ArgumentParser(
description=
"Make sam file compatible with tools counting on a splicemap format sam file."
)
parser.add_argument('in_sam',
help="FILENAME of sam file, or '-' for STDIN")
args = parser.parse_args()
inf = sys.stdin
if args.in_sam != '-':
inf = open(args.in_sam)
for line in inf:
line = line.rstrip()
if SamBasics.is_header(line):
print line
continue
f = line.rstrip().split("\t")
e = SamBasics.sam_line_to_dictionary(line)
if SamBasics.check_flag(e['flag'], 4):
continue # skip the unmapped reads
if SamBasics.check_flag(e['flag'], 16):
f[1] = "16"
else:
f[1] = "0"
f[4] = "0"
f[6] = "*"
f[7] = "0"
f[8] = "0"
print "\t".join(f)
def main():
parser = argparse.ArgumentParser(description="Make sam file compatible with tools counting on a splicemap format sam file.")
parser.add_argument('in_sam',help="FILENAME of sam file, or '-' for STDIN")
args = parser.parse_args()
inf = sys.stdin
if args.in_sam != '-':
inf = open(args.in_sam)
for line in inf:
line = line.rstrip()
if SamBasics.is_header(line):
print line
continue
f = line.rstrip().split("\t")
e = SamBasics.sam_line_to_dictionary(line)
if SamBasics.check_flag(e['flag'],4):
continue # skip the unmapped reads
if SamBasics.check_flag(e['flag'],16):
f[1] = "16"
else:
f[1] = "0"
f[4] = "0"
f[6] = "*"
f[7] = "0"
f[8] = "0"
print "\t".join(f)
def make_exons(args, thread_index, thread_count):
is_sam = True
if re.search('\.bam$', args.sam_file):
is_sam = False
stag = ''
if is_sam: stag = '-S'
cmd = 'samtools view -F 4 ' + stag + ' ' + args.sam_file
spcf = SamBasics.SAMtoPSLconversionFactory()
if args.reference_genome:
spcf.set_genome(args.reference_genome)
sampipe = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE)
fname = args.tempdir + '/bedpart.' + str(thread_index) + '.bed'
of = open(fname, 'w')
z = 0
with sampipe.stdout as inf:
for line in inf:
z += 1
if z % thread_count != thread_index: continue
line = line.rstrip()
if SamBasics.is_header(line):
continue
d = SamBasics.sam_line_to_dictionary(line)
strand = '+'
if SamBasics.check_flag(d['flag'], 16):
strand = '-'
seqs = []
sequence = d['seq']
seqs.append([d['qname'], d['rname'], strand, d['pos'], d['cigar']])
m = re.search('XA:Z:(\S+)', line)
if m and args.use_secondary_alignments:
e = m.group(1)
secondaries = e.rstrip(";").split(";")
for secondary in secondaries:
m1 = re.match('([^,]+),([+-])(\d+),([^,]+)', secondary)
if not m1:
sys.stderr.write("strange secondary format " +
secondary + "\n")
sys.exit()
seqs.append([
d['qname'],
m1.group(1),
m1.group(2),
int(m1.group(3)),
m1.group(4)
])
#p.apply_async(get_exons_from_seqs,[seqs,d,spcf])
exons = get_exons_from_seqs(seqs, d, spcf)
of.write(exons)
#return exons
of.close()
def make_exons(args,thread_index,thread_count):
is_sam = True
if re.search('\.bam$',args.sam_file):
is_sam = False
stag = ''
if is_sam: stag = '-S'
cmd = 'samtools view -F 4 '+stag+' '+args.sam_file
spcf = SamBasics.SAMtoPSLconversionFactory()
if args.reference_genome:
spcf.set_genome(args.reference_genome)
sampipe = subprocess.Popen(cmd,shell=True,stdout=subprocess.PIPE)
fname = args.tempdir+'/bedpart.'+str(thread_index)+'.bed'
of = open(fname,'w')
z = 0
with sampipe.stdout as inf:
for line in inf:
z+=1
if z%thread_count != thread_index: continue
line = line.rstrip()
if SamBasics.is_header(line):
continue
d = SamBasics.sam_line_to_dictionary(line)
strand = '+'
if SamBasics.check_flag(d['flag'],16):
strand = '-'
seqs = []
sequence = d['seq']
seqs.append([d['qname'], d['rname'], strand, d['pos'], d['cigar']])
m = re.search('XA:Z:(\S+)',line)
if m and args.use_secondary_alignments:
e = m.group(1)
secondaries = e.rstrip(";").split(";")
for secondary in secondaries:
m1 = re.match('([^,]+),([+-])(\d+),([^,]+)',secondary)
if not m1:
sys.stderr.write("strange secondary format "+secondary+"\n")
sys.exit()
seqs.append([d['qname'], m1.group(1),m1.group(2),int(m1.group(3)),m1.group(4)])
#p.apply_async(get_exons_from_seqs,[seqs,d,spcf])
exons = get_exons_from_seqs(seqs,d,spcf)
of.write(exons)
#return exons
of.close()
def main():
parser = argparse.ArgumentParser(description="Read a sam file and output a bed file in the format of junction_color.bed")
parser.add_argument('-o','--output',help='FILENAME is output')
parser.add_argument('--min_intron_size',type=int,default=68,help='minimum intron size')
parser.add_argument('infile',help='FILENAME of sam file or "-" for STDIN')
parser.add_argument('reference_genome',help='FILENAME of the reference genome')
args = parser.parse_args()
# get our reference genome
sys.stderr.write("reading reference genome\n")
#g = SequenceBasics.read_fasta_into_hash(args.reference_genome)
g = FastaData(open(args.reference_genome).read())
sys.stderr.write("finished reading reference genome\n")
inf = sys.stdin
read_mapping_count = {}
junctions = {}
if args.infile != '-':
inf = open(args.infile)
sys.stderr.write("reading through sam file\n")
zall = 0
zn = 0
while True:
line = inf.readline()
if not line: break
line = line.rstrip()
if SamBasics.is_header(line): continue
d = SamBasics.sam_line_to_dictionary(line)
chrom = d['rname']
if chrom =='*': continue
if chrom not in g.keys():
sys.stderr.write("WARNING: "+chrom+" not in reference, skipping\n")
continue
mate = 'U'
if SamBasics.check_flag(d['flag'],int('0x4',16)): #check if its unmapped
continue # we can ignore the unmapped things for now
if SamBasics.check_flag(d['flag'],int('0x40',16)):
mate = 'L'
elif SamBasics.check_flag(d['flag'],int('0x80',16)):
mate = 'R'
actual_read = d['qname']+"\t"+mate
if actual_read not in read_mapping_count:
read_mapping_count[actual_read] = 0
read_mapping_count[actual_read] += 1
has_intron = 0
start_loc = d['pos']
current_loc = start_loc
bounds = []
for i in range(0,len(d['cigar_array'])):
ce = d['cigar_array'][i]
if ce['op'] == 'N' and ce['val'] >= args.min_intron_size:
has_intron = 1
lbound = current_loc # should be the intron start base index-1
current_loc += ce['val']
rbound = current_loc # should be the second exon start base index-1
right_size = d['cigar_array'][i+1]['val']
bounds.append([lbound,rbound,right_size])
elif ce['op'] == 'D':
current_loc += ce['val']
elif re.match('[=XMSHP]',ce['op']):
current_loc += ce['val']
if has_intron == 0: continue # there are no splices to report here
#print actual_read
#print d['cigar']
#print d
#print start_loc
#print bounds
for bound in bounds:
zall += 1
intronflank = g[chrom][bound[0]-1:bound[0]+1].upper() + '-' + \
g[chrom][bound[1]-3:bound[1]-1].upper()
strand = ''
if is_canon(intronflank): # its a positive strand
strand = '+'
elif is_revcanon(intronflank): # its a negative strand
strand = '-'
else:
# We can't deal with the non-canonical splice sorry
zn += 1
sys.stderr.write("WARNING skipping non-canonical splice ("+str(zn)+"/"+str(zall)+")\r")
continue
# If we are still in we have successfully found a splice
out_chrom = chrom
out_start = bound[0]-51
out_end = bound[1]+49
out_name = '*' # this will be done later
out_score = 50
out_strand = strand
out_thickStart = out_start
out_thickEnd = out_end
out_rgb = '0,0,0'
out_block_count = 2
out_block_sizes = '50,50'
out_block_starts = '0,'+str(bound[1]-bound[0]+50)
bed = []
bed.append(out_chrom)
bed.append(str(out_start))
bed.append(str(out_end))
bed.append(out_name)
bed.append(str(out_score))
bed.append(out_strand)
bed.append(str(out_thickStart))
bed.append(str(out_thickEnd))
bed.append(out_rgb)
bed.append(str(out_block_count))
bed.append(out_block_sizes)
bed.append(out_block_starts)
entry = "\t".join(bed)
if entry not in junctions:
junctions[entry] = {}
junctions[entry]['reads'] = set()
junctions[entry]['positions'] = set()
junctions[entry]['right_sizes'] = set()
junctions[entry]['reads'].add(actual_read)
junctions[entry]['positions'].add(d['pos'])
junctions[entry]['right_sizes'].add(bound[2])
sys.stderr.write("\n")
sys.stderr.write("finished reading sam\n")
of = sys.stdout
if args.output:
of = open(args.output,'w')
if len(junctions) > 0: # if we have stuff lets print a header
of.write("track\tname=junctions\tdescription=\"SpliceMap junctions\" itemRgb=\"On\"\n")
for entry in junctions:
nR = len(junctions[entry]['reads'])
width = max(junctions[entry]['right_sizes'])-min(junctions[entry]['right_sizes'])
nNR = len(junctions[entry]['positions'])
nUR = 0
nMR = 0
for read in junctions[entry]['reads']:
if read_mapping_count[read] == 1:
nUR += 1
elif read_mapping_count[read] > 1:
nMR += 1
else:
sys.stderr.write("ERROR: nonsense read count\n")
return
name = '('+str(nR)+')['+str(width)+'_'+str(nNR)+']('+str(nUR)+'/'+str(nMR)+')'
bed = entry.split("\t")
bed[3] = name
of.write("\t".join(bed)+"\n")
def main():
parser = argparse.ArgumentParser(
description="Convert a sam file into a psl file")
parser.add_argument('--genome',
help="FASTA input file of reference genome")
parser.add_argument('--get_secondary_alignments',
action='store_true',
help="Report SA:Z secondary alignments as well")
parser.add_argument('--get_alternative_alignments',
action='store_true',
help="Report XA:Z alternative alignments as well")
parser.add_argument(
'--get_all_alignments',
action='store_true',
help="Report SA:Z and XA:Z alternative alignments as well")
parser.add_argument('--give_unique_names',
action='store_true',
help="Output query names will be unique.")
group = parser.add_mutually_exclusive_group()
group.add_argument(
'--output_fasta',
help=
"FILENAME to save an outgoing fasta. Only works for primary alignments."
)
group.add_argument(
'--output_fastq',
help=
"FILENAME to save an outgoing fastq. Only works for primary alignments."
)
parser.add_argument('infile', help="FILENAME input file or '-' for STDIN")
parser.add_argument('-o',
'--output',
help="FILENAME for the output, STDOUT if not set.")
args = parser.parse_args()
if (args.output_fasta
or args.output_fastq) and (args.get_secondary_alignments
or args.get_alternative_alignments
or args.get_all_alignments):
sys.stderr.write(
"ERROR, can only output the fastq/fasta if we are doing primary alignments only.\n"
)
sys.exit()
inf = sys.stdin
if args.infile != '-':
inf = open(args.infile)
of = sys.stdout
if args.output:
of = open(args.output, 'w')
spcf = SamBasics.SAMtoPSLconversionFactory()
if args.genome: spcf.set_genome(args.genome)
off = None
if args.output_fasta:
off = open(args.output_fasta, 'w')
if args.output_fastq:
off = open(args.output_fastq, 'w')
z = 0
for line in inf:
line = line.rstrip()
if SamBasics.is_header(line):
spcf.read_header_line(line)
continue
# We have a line to convert
psl = spcf.convert_line(line)
if psl:
pobj = PSL(psl)
z += 1
if args.give_unique_names:
pobj.entry['qName'] = 'Q' + str(z)
of.write(pobj.get_line() + "\n")
if args.output_fastq or args.output_fasta:
sam = SamBasics.SAM(line)
sequence = sam.value('seq').upper()
quality = sam.value('qual')
if sam.check_flag(16):
sequence = rc(sam.value('seq').upper())
quality = sam.value('qual')[::-1]
if args.output_fasta:
off.write(">" + pobj.value('qName') + "\n" + sequence +
"\n")
elif args.output_fastq:
if len(sequence) == len(quality):
off.write("@" + pobj.value('qName') + "\n" + sequence +
"\n" + "+\n" + quality + "\n")
else:
sys.stderr.write("ERROR: sequence " + sequence +
" length (" + str(len(sequence)) +
") doesnt match quality " + quality +
" length (" + str(len(quality)) +
")\n")
sys.exit()
# Lets look for secondary alignments to convert
if args.get_secondary_alignments or args.get_all_alignments:
secondary_alignments = SamBasics.get_secondary_alignments(
line.rstrip())
for samline in secondary_alignments:
psl = spcf.convert_line(samline)
if psl:
#print "\nsecondary"
#print samline
z += 1
pobj = PSL(psl)
if args.give_unique_names:
pobj.entry['qName'] = 'Q' + str(z)
of.write(pobj.get_line() + "\n")
if args.get_alternative_alignments or args.get_all_alignments:
alternative_alignments = SamBasics.get_alternative_alignments(
line.rstrip())
for samline in alternative_alignments:
psl = spcf.convert_line(samline)
if psl:
#print "\nsecondary"
#print samline
z += 1
pobj = PSL(psl)
if args.give_unique_names:
pobj.entry['qName'] = 'Q' + str(z)
of.write(pobj.get_line() + "\n")
inf.close()
of.close()
def main():
parser = argparse.ArgumentParser(description="Convert a sam file into a psl file")
parser.add_argument('--genome',help="FASTA input file of reference genome")
parser.add_argument('--get_secondary_alignments',action='store_true',help="Report SA:Z secondary alignments as well")
parser.add_argument('--get_alternative_alignments',action='store_true',help="Report XA:Z alternative alignments as well")
parser.add_argument('--get_all_alignments',action='store_true',help="Report SA:Z and XA:Z alternative alignments as well")
parser.add_argument('--give_unique_names',action='store_true',help="Output query names will be unique.")
group = parser.add_mutually_exclusive_group()
group.add_argument('--output_fasta',help="FILENAME to save an outgoing fasta. Only works for primary alignments.")
group.add_argument('--output_fastq',help="FILENAME to save an outgoing fastq. Only works for primary alignments.")
parser.add_argument('infile',help="FILENAME input file or '-' for STDIN")
parser.add_argument('-o','--output',help="FILENAME for the output, STDOUT if not set.")
args = parser.parse_args()
if (args.output_fasta or args.output_fastq) and (args.get_secondary_alignments or args.get_alternative_alignments or args.get_all_alignments):
sys.stderr.write("ERROR, can only output the fastq/fasta if we are doing primary alignments only.\n")
sys.exit()
inf = sys.stdin
if args.infile != '-':
inf = open(args.infile)
of = sys.stdout
if args.output:
of = open(args.output,'w')
spcf = SamBasics.SAMtoPSLconversionFactory()
if args.genome: spcf.set_genome(args.genome)
off = None
if args.output_fasta:
off = open(args.output_fasta,'w')
if args.output_fastq:
off = open(args.output_fastq,'w')
z = 0
for line in inf:
line = line.rstrip()
if SamBasics.is_header(line):
spcf.read_header_line(line)
continue
# We have a line to convert
psl = spcf.convert_line(line)
if psl:
pobj = PSL(psl)
z += 1
if args.give_unique_names:
pobj.entry['qName'] = 'Q'+str(z)
of.write(pobj.get_line()+"\n")
if args.output_fastq or args.output_fasta:
sam = SamBasics.SAM(line)
sequence = sam.value('seq').upper()
quality = sam.value('qual')
if sam.check_flag(16):
sequence = rc(sam.value('seq').upper())
quality = sam.value('qual')[::-1]
if args.output_fasta:
off.write(">"+pobj.value('qName')+"\n"+sequence+"\n")
elif args.output_fastq:
if len(sequence) == len(quality):
off.write("@"+pobj.value('qName')+"\n"+sequence+"\n"+"+\n"+quality+"\n")
else:
sys.stderr.write("ERROR: sequence "+sequence+" length ("+str(len(sequence))+") doesnt match quality "+quality+" length ("+str(len(quality))+")\n")
sys.exit()
# Lets look for secondary alignments to convert
if args.get_secondary_alignments or args.get_all_alignments:
secondary_alignments = SamBasics.get_secondary_alignments(line.rstrip())
for samline in secondary_alignments:
psl = spcf.convert_line(samline)
if psl:
#print "\nsecondary"
#print samline
z += 1
pobj = PSL(psl)
if args.give_unique_names:
pobj.entry['qName'] = 'Q'+str(z)
of.write(pobj.get_line()+"\n")
if args.get_alternative_alignments or args.get_all_alignments:
alternative_alignments = SamBasics.get_alternative_alignments(line.rstrip())
for samline in alternative_alignments:
psl = spcf.convert_line(samline)
if psl:
#print "\nsecondary"
#print samline
z += 1
pobj = PSL(psl)
if args.give_unique_names:
pobj.entry['qName'] = 'Q'+str(z)
of.write(pobj.get_line()+"\n")
inf.close()
of.close()