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 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 main():
# parse options
options = docopt(__doc__, version=__version__)
if options['--seq']:
if not os.path.isfile(options['--seq']):
sys.exit('Error: wrong seq file!')
seq = os.path.abspath(options['--seq'])
seq_flag = True
else:
seq = None
seq_flag = False
fa = check_fasta(options['--genome'])
chrom = options['--chrom']
site = int(options['--site'])
strand = '+' if options['--strand'] == '1' else '-'
rlen = int(options['--read-length'])
alen = int(options['--region-length'])
clen = int(options['--check-region-length'])
thread = options['--thread']
skip_flag = options['--skip-alignment']
# check output directory
if not skip_flag: # not skip alignment
out_dir = create_dir(options['<out_dir>'])
else: # skip alignment
out_dir = check_dir(options['<out_dir>'])
# build index for sgRNA
index_path, offset = build_index(fa, chrom, site, strand, rlen, thread,
out_dir, seq, seq_flag)
if not skip_flag: # not skip alignment
# deal with reads file
reads = tempfile.NamedTemporaryFile(mode='w+')
if options['-R']:
fq_lst = options['-R'].split(',')
convert_read(reads, single=fq_lst)
else:
fq1_lst = options['-1'].split(',')
fq2_lst = options['-2'].split(',')
convert_read(reads, fq1=fq1_lst, fq2=fq2_lst)
reads.seek(0)
read_path = reads.name
# mapped reads with bowtie2
bam = bowtie2_align(index_path, read_path, thread, out_dir)
# remove tempfile
reads.close()
else:
bam = os.path.join(out_dir, 'cs.bam')
# fetch cleavage site reads
fetch_reads(index_path, offset, alen, clen, bam, out_dir)
def fetch_reads(index, offset, alen, clen, bam, out_dir):
cs = os.path.join(out_dir, 'cs_region.txt')
count = os.path.join(out_dir, 'cs_count.txt')
with open(cs, 'w') as cs_out, open(count, 'w') as count_out:
cs_out.write(' ' * alen + 'sgRNA|PAM|\n')
index_fa = check_fasta(index).fetch('sgRNA_region')
cs_out.write('Reference: ' + index_fa[offset - alen:offset + alen] +
'\n')
count_out.write('Reference: ' + index_fa[offset - clen:offset + clen] +
'\n')
cs_count = defaultdict(int)
bam_f = pysam.AlignmentFile(bam, 'rb')
for read in bam_f:
seq = read.query_sequence
start = read.reference_start
pos = 0
align = insert = ' ' * start
for tag, tlen in read.cigartuples:
if tag == 0: # M
align += seq[pos:pos + tlen]
insert += ' ' * tlen
pos += tlen
elif tag == 1: # I
insert += seq[pos:pos + tlen]
pos += tlen
else: # D
align += '*' * tlen
cs_out.write('Alignment: ' + align[offset - alen:offset + alen] +
'\n')
cs_out.write('Insertion: ' + insert[offset - alen:offset + alen] +
'\n')
cs_id = 'Reads : ' + align[offset - clen:offset + clen]
cs_id += '\t%d\n'
cs_id += 'Indel :' + insert[offset - clen:offset + clen] + '\n'
cs_count[cs_id] += 1
for cs_id in cs_count:
count_out.write(cs_id % cs_count[cs_id])