def measure_length_in_fastq(fastq, split_length, coord1, coord2):
"""Measure the distances between 2 coordinates in rDNA for a set of reads.
This module takes a .fastq file that contains many multiple rDNA containing
reads and measures the length for each of them and returns the result.
Coordinate1 should be smaller than coordinate 2.
Args:
fastq (str): fastq filename
split_length (int): split length when the read is mapped to rDNA
coordinate1 (int): coordinate in rDNA
coordinate2 (int): coordinate in rDNA
Returns:
result: list of (header, lisf of measured distance between 2 coords)
"""
with open(fastq) as f:
result = []
for data in itertools.zip_longest(*[iter(f)] * 4):
header = data[0].strip()
read = data[1].strip()
quality = data[3].strip()
make_temp_fastq(split_length, header, read, quality)
subprocess.run(
'bwa mem -M -x ont2d -t 5 /home/yutaro/nanopore/'
'clive/rDNA_index/humRibosomal.fa temp_fastq.fastq'
' > temp_sam.sam',
shell=True,
stdout=FNULL,
stderr=subprocess.STDOUT)
with open('temp_sam.sam') as samf:
samdata = samf.readlines()[2:]
if is_rDNAs_healthy(samdata):
dist = measure_length_main(header, read, coord1, coord2)
result.append((header.split()[0], dist))
return result
def analyze_fastq_by_header(fastq, header_id, split_length):
"""Find a read from a fastq file and analyze its structure.
Args:
fastq (str): filename
header_id (str): header id
split_length (int): split length
Returns:
TRs: TRs
"""
with open(fastq) as f:
for line in f:
if header_id in line:
header = line
read = f.readline()
f.readline()
quality = f.readline()
make_temp_fastq(split_length, header, read, quality)
FNULL = open(os.devnull, 'w')
subprocess.run(
'bwa mem -M -x ont2d -t 5 /home/yutaro/nanopore/'
'clive/rDNA_index/humRibosomal.fa '
'temp_files/temp_fastq.fastq > temp_files/'
'single_split_mapped.sam',
shell=True,
stdout=FNULL,
stderr=subprocess.STDOUT)
break
TRs = analyze_split_reads('temp_files/single_split_mapped.sam',
split_length)
return TRs
def analyze_all_fastq_file(fastq, split_length, length_cutoff):
"""Perform TR analysis for a fastq and output a summarized list.
For each element in set_of_TRs, true TRs is element[1]!
Args:
fastq (str): fastq filename
split_length (int): split length
length_cutoff (int): reads shorter than this are omitted
Returns:
set_of_TRs: each TRs obtained from analyze_split_reads are paired with
header as [header, TRs] and summarized as set_of_TRs
"""
# this function read a fastq file and split each read and then map them to rDNA.
# Each mapped read is analyzed by analyze_split_reads to make TRs.
count = 0
set_of_TRs = []
with open(fastq) as f:
for n, line in enumerate(f):
if n % 4 == 0:
header = line.strip()
if n % 4 == 1:
read = line.strip()
if n % 4 == 3:
quality = line.strip()
if len(quality) < length_cutoff:
continue
else:
count += 1
make_temp_fastq(split_length, header, read, quality)
FNULL = open(os.devnull, 'w')
subprocess.run(
'bwa mem -M -x ont2d -t 5 /home/yutaro/nanopore/clive/rDNA_index/humRibosomal.fa temp_files/temp_fastq.fastq > temp_files/temp_sam.sam',
shell=True,
stdout=FNULL,
stderr=subprocess.STDOUT)
TRs = analyze_split_reads('temp_files/temp_sam.sam',
split_length)
set_of_TRs.append([header, TRs])
return set_of_TRs
def find_boundaries_from_fastq(fastq, split_length):
"""Find boundary containing reads from a .fastq file.
Args:
fastq (str): filename
split_length (split_length): split_length
Returns:
boundaries: list of (boundary coordinate, boundary rDNA coordinate,
direction, side of non rDNA, read, header)
"""
with open(fastq) as f:
boundaries = []
for n, each_fastq in enumerate(itertools.zip_longest(*[iter(f)] * 4)):
header = each_fastq[0].strip()
read = each_fastq[1].strip()
if len(read) < 40000:
continue
quality = each_fastq[3].strip()
make_temp_fastq(split_length, header, read, quality)
subprocess.run(
'bwa mem -M -x ont2d -t 5 '
'/home/yutaro/nanopore/clive/rDNA_index/'
'humRibosomal.fa temp_files/temp_fastq.fastq > '
'temp_files/temp_sam.sam',
shell=True,
stdout=FNULL,
stderr=subprocess.STDOUT)
# rDNA_coordinate=1 when using find_true_boundary2
temp_boundary = find_end_reads('temp_files/temp_sam.sam',
split_length,
rDNA_coordinate=1)
if temp_boundary:
header = header.split()[0]
boundary = int(temp_boundary[0])
direction = temp_boundary[1]
side = temp_boundary[2]
if direction == '+':
if side == 'right':
bound_seq = read[boundary:boundary + 10000]
with open('boundary_seq1.fa', 'a') as fw:
fw.write('>' + header + '\n')
fw.write(bound_seq + '\n\n')
else:
with open('boundary_seq2.fa', 'a') as fw:
fw.write('>' + header + '\n')
fw.write(bound_seq + '\n\n')
else:
bound_seq = read[boundary - 10000:boundary]
revcom = str(Seq(bound_seq).reverse_complement())
if side == 'left':
with open('boundary_seq1.fa', 'a') as fw:
fw.write('>' + header + '\n')
fw.write(revcom + '\n\n')
else:
with open('boundary_seq2.fa', 'a') as fw:
fw.write('>' + header + '\n')
fw.write(revcom + '\n\n')
plot_read_structure(header,
split_length,
savename='end_reads/' + header + '.png',
title=str(temp_boundary[0]))
continue
true_boundary = find_true_boundary2(header, read, quality,
temp_boundary)
if true_boundary:
boundaries.append(
(true_boundary[0], true_boundary[1], temp_boundary[1],
temp_boundary[2], read, header.split()[0]))
return boundaries
id2scores = {}
with h5py.File('megalodon_results/basecalls.modified_base_scores.hdf5') as f:
for read_id in rDNA_read_ids:
mc_data = np.array(f['Reads'][read_id])[:, 1]
ma_data = np.array(f['Reads'][read_id])[:, 0]
read = id2read[read_id]
cpgs = find_cpg(read)
scores_c = []
for i in cpgs:
scores_c.append([i, mc_data[i]])
id2scores[read_id] = scores_c
split_length = 200
for read_id in rDNA_read_ids:
read = id2read[read_id]
search_rDNA_reads.make_temp_fastq(split_length, read_id, read,
'J' * len(read))
FNULL = open(os.devnull, 'w')
subprocess.run(
'bwa mem -M -x ont2d -t 5 reference/rDNA_for_cas9_no_margin.fasta '
'temp_files/temp_fastq.fastq > temp_files/'
'single_split_mapped.sam',
shell=True,
stdout=FNULL,
stderr=subprocess.STDOUT)
lc = plot_read_structure('test', split_length, 9400)
read_len = len(read)
fig = plt.figure()
plt.subplots_adjust(left=0.2)
ax = fig.add_subplot()
x = []
y = []