def fetch_read2(bam, chrom):
read2_lst = {}
for read in bam.fetch(chrom):
# not read1 or secondary alignment
if read.is_read1 or read.is_secondary:
continue
if not read.is_proper_pair: # not proper pair
continue
if read.get_tag('NH') != 1: # not unique read
continue
chrom = read.reference_name
mate_chrom = read.next_reference_name
if chrom != mate_chrom: # not same chromosome
continue
if not read.is_reverse and read.mate_is_reverse:
strand = '+'
mate_strand = '-'
elif read.is_reverse and not read.mate_is_reverse:
strand = '-'
mate_strand = '+'
else:
continue
mate_pos = str(read.next_reference_start)
name = read.query_name
start = str(read.reference_start)
end = str(read.reference_end)
if strand == '+':
barcode = dna_to_rna(read.query_sequence[:15])
else:
barcode = dna_to_rna(read.query_sequence[-15:], strand=strand)
read_id = '\t'.join([name, mate_chrom, mate_pos, mate_strand])
read2_lst[read_id] = [chrom, start, end, barcode]
return read2_lst
def parse_intron(options, chrom, start, end, strand, intron_info):
# fetch fasta
fa = check_fasta(options['--genome'])
intron_fa = dna_to_rna(fa.fetch(chrom, start, end), strand)
# load matrix
matrix3 = load_matrix3()
# parse options
phastcons_f = pyBigWig.open(options['--bigwig'])
min_distance = int(options['--min-distance'])
min_score = float(options['--min-score'])
min_phastcons = float(options['--min-phastcons'])
# start to parse rs sites
rs_list = []
for m in re.finditer('AGGT', intron_fa):
if strand == '+':
pos = start + m.start() + 2
left_dist, right_dist, dist_flag = cal_distance(
pos, start, end, min_distance)
if not dist_flag: # not enough distance
continue
ss3_seq = dna_to_rna(fa.fetch(chrom, pos - 20, pos + 3))
if ss3_seq.find('N') != -1: # ensure there is no N
continue
ss3, score_flag = cal_score(ss3_seq, matrix3, min_score)
if not score_flag: # not high score
continue
else:
pos = end - m.start() - 2
left_dist, right_dist, dist_flag = cal_distance(
pos, start, end, min_distance)
if not dist_flag: # not enough distance
continue
ss3_seq = dna_to_rna(fa.fetch(chrom, pos - 3, pos + 20),
strand='-')
if ss3_seq.find('N') != -1: # ensure there is no N
continue
ss3, score_flag = cal_score(ss3_seq, matrix3, min_score)
if not score_flag: # not high score
continue
phastcons = phastcons_f.stats(chrom, pos - 2, pos + 2)[0]
if phastcons is None or phastcons < min_phastcons: # not conserved
continue
rs_feature = '%d|%d|%d|%f|%f' % (pos, left_dist, right_dist, ss3,
phastcons)
rs_list.append(rs_feature)
if rs_list:
return (intron_info, rs_list)
else:
return (None, None)
def parse_intron(options, chrom, start, end, strand, intron_info):
# fetch fasta
fa = check_fasta(options['--genome'])
intron_fa = dna_to_rna(fa.fetch(chrom, start, end), strand)
# parse options
phastcons_f = pyBigWig.open(options['--bigwig'])
min_distance = int(options['--min-distance'])
# start to parse rs sites
rs_list = defaultdict(list)
for i in product('ATCG', repeat=4):
motif = ''.join(i)
rs_list[motif].append('x')
for m in re.finditer(motif, intron_fa):
if strand == '+':
pos = start + m.start() + 2
left_dist, right_dist, dist_flag = cal_distance(pos, start,
end,
min_distance)
if not dist_flag: # not enough distance
continue
else:
pos = end - m.start() - 2
left_dist, right_dist, dist_flag = cal_distance(pos, start,
end,
min_distance)
if not dist_flag: # not enough distance
continue
phastcons = phastcons_f.stats(chrom, pos - 2, pos + 2)[0]
if phastcons is None: # no conservation score
continue
rs_list[motif].append(phastcons)
intron_length = end - start - 2 * min_distance
return(rs_list, intron_length)
def parse_intron(options, chrom, start, end, strand, intron_info):
# fetch fasta
fa = check_fasta(options['--genome'])
intron_fa = dna_to_rna(fa.fetch(chrom, start, end), strand)
# parse options
motif = options['-m']
phastcons_f = pyBigWig.open(options['--bigwig'])
min_distance = int(options['--min-distance'])
min_phastcons = float(options['--min-phastcons'])
# start to parse rs sites
rs_list = []
for m in re.finditer(motif, intron_fa):
if strand == '+':
pos = start + m.start() + 2
left_dist, right_dist, dist_flag = cal_distance(
pos, start, end, min_distance)
if not dist_flag: # not enough distance
continue
else:
pos = end - m.start() - 2
left_dist, right_dist, dist_flag = cal_distance(
pos, start, end, min_distance)
if not dist_flag: # not enough distance
continue
phastcons = phastcons_f.stats(chrom, pos - 2, pos + 2)[0]
if phastcons is None or phastcons < min_phastcons: # not conserved
continue
rs_feature = '%d|%d|%d|%f' % (pos, left_dist, right_dist, phastcons)
rs_list.append(rs_feature)
if rs_list:
return (intron_info, rs_list)
else:
return (None, None)
def build_index(fa, chrom, site, strand, rlen, thread, out_dir, seq, seq_flag):
print('Build index...')
if strand == '+':
start = site - (rlen - 10)
end = site + (rlen - 20)
offset = rlen - 10
else:
start = site - (rlen - 20)
end = site + (rlen - 10)
offset = rlen - 20
index_path = os.path.join(out_dir, 'sgRNA.fa')
if seq_flag:
os.symlink(seq, index_path)
else:
# fetch sgRNA region sequence
with open(index_path, 'w') as out:
out.write('>sgRNA_region\n')
out.write(
dna_to_rna(fa.fetch(chrom, start, end), strand=strand) + '\n')
# build index
if which('bowtie2-build'):
command = 'bowtie2-build -q --threads %s %s %s'
command = command % (thread, index_path, index_path)
run_command(command, 'Error: cannot build index for sgRNA!')
else:
sys.exit('Error: no bowtie2-build installed!')
return index_path, offset