-
Notifications
You must be signed in to change notification settings - Fork 0
/
enrichmenTE_detect.py
135 lines (114 loc) · 4.46 KB
/
enrichmenTE_detect.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
#!/usr/bin/env python3
import os
import subprocess
from glob import glob
from enrichmenTE_utility import *
"""
this module takes family and primer set multiplexed pair end illumina data (if data is primer set/family non-multiplexed, a list of R1/R2 can be provided) and output reference/non-reference TEs for six focal TE families: 'copia', '1731', 'roo', 'mdg3', 'mdg1', '297'.
"""
def detect(
prefix,
read1,
read2,
reference,
outdir,
depth_config,
ref_te_bed,
window,
tsd_max,
gap_max,
filter_region,
thread,
):
families = ["1731", "297", "copia", "mdg1", "mdg3", "roo"]
# process input read list
read1_list = read1.split(",")
read2_list = read2.split(",")
input_files = read1 + "," + read2
input_files = input_files.replace(" ", "")
# unzip and merge input files for R1 and R2
fq1 = outdir + "/" + prefix + ".R1.fastq"
with open(fq1, "w") as output:
for read1 in read1_list:
subprocess.call(["gunzip", "-c", read1], stdout=output)
fq2 = outdir + "/" + prefix + ".R2.fastq"
with open(fq2, "w") as output:
for read2 in read2_list:
subprocess.call(["gunzip", "-c", read2], stdout=output)
# align R2 to masked augmented reference genome
bam_r2 = outdir + "/" + prefix + ".R2.bam"
make_bam(fq2, reference, str(thread), bam_r2)
# align R1 to masked augmented reference genome
bam_r1 = outdir + "/" + prefix + ".R1.bam"
make_bam(fq1, reference, str(thread), bam_r1)
# clean fastq files
os.remove(fq1)
os.remove(fq2)
# demultiplex BAM file by family and primer set
sets = ["set1", "set2"]
for family in families:
family_dir = os.path.join(outdir, family)
mkdir(family_dir)
for set in sets:
group_name = ".".join([prefix, family, set])
read_list = family_dir + "/" + group_name + ".txt"
extract_reads(bam_r2, family, set, read_list)
# use read ID to filter read 1 bam file by family and set
if os.path.isfile(read_list) and os.stat(read_list).st_size != 0:
bam_out = family_dir + "/" + group_name + ".R1.bam"
filter_bam(bam_r1, read_list, bam_out)
else:
print("family with no R2 alignment for " + set + ": " + family)
os.remove(read_list)
# for each family, generate reference/non-reference TE predictions and report summary
for family in families:
family_dir = os.path.join(outdir, family)
bam_set1 = family_dir + "/" + prefix + "." + family + ".set1.R1.bam"
bam_set2 = family_dir + "/" + prefix + "." + family + ".set2.R1.bam"
bed_set1 = bam_set1.replace("bam", "cluster.bed")
bed_set2 = bam_set2.replace("bam", "cluster.bed")
if os.path.isfile(bam_set1):
get_cluster(
bam_set1,
bed_set1,
config=depth_config,
window=window,
set="set1",
family=family,
)
os.remove(bam_set1)
if os.path.isfile(bam_set2):
get_cluster(
bam_set2,
bed_set2,
config=depth_config,
window=window,
set="set2",
family=family,
)
os.remove(bam_set2)
if os.path.isfile(bed_set1) and os.path.isfile(bed_set2):
get_nonref(bed_set1, bed_set2, family_dir, family, tsd_max, gap_max)
# gather non-reference TE predictions
nonref_tmp = outdir + "/" + prefix + ".nonref.tmp.bed"
pattern = "/*/*.nonref.bed"
bed_files = glob(outdir + pattern, recursive=True)
genome = get_genome_file(reference)
merge_bed(
bed_in=bed_files, bed_out=nonref_tmp, genome=genome, filter_method="overlap"
)
# remove non-reference TE predictions that overlap reference TE annotations
nonref_filter1 = outdir + "/" + prefix + ".nonref.filter1.bed"
with open(nonref_filter1, "w") as output:
subprocess.call(
["bedtools", "intersect", "-v", "-a", nonref_tmp, "-b", ref_te_bed],
stdout=output,
)
# filter non-reference TE predictions in target regions
nonref_filter2 = outdir + "/" + prefix + ".nonref.filter2.bed"
with open(nonref_filter2, "w") as output:
subprocess.call(
["bedtools", "intersect", "-a", nonref_filter1, "-b", filter_region],
stdout=output,
)
return nonref_filter2