-
Notifications
You must be signed in to change notification settings - Fork 1
/
geniso
executable file
·137 lines (117 loc) · 5.01 KB
/
geniso
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
136
137
#!/usr/bin/env python3
import argparse
import isoform
## Command Line Interface ##
parser = argparse.ArgumentParser(
description='Alternative isoform generator')
parser.add_argument('fasta', type=str, metavar='<file>',
help='input fasta file (reads only the first sequence if multi-fasta)')
parser.add_argument('--min_intron', required=False, type=int, default=35,
metavar='<int>', help='minimum length of intron [%(default)i]')
parser.add_argument('--min_exon', required=False, type=int, default=25,
metavar='<int>', help='minimum length exon [%(default)i]')
parser.add_argument('--max_splice', required=False, type=int, default=3,
metavar='<int>', help='maximum number of introns [%(default)i]')
parser.add_argument('--flank', required=False, type=int, default=99,
metavar='<int>', help='genomic flank on each side [%(default)i]')
parser.add_argument('--dpwm', required=False, type=str, metavar='<file>',
help='position weight matrix for donor site [%(default)s]')
parser.add_argument('--apwm', required=False, type=str, metavar='<file>',
help='position weight matrix for acceptor site [%(default)s]')
parser.add_argument('--emm', required=False, type=str, metavar='<file>',
help='markov model for exon sequence [%(default)s]')
parser.add_argument('--imm', required=False, type=str, metavar='<file>',
help='markov model for intron sequence [%(default)s]')
parser.add_argument('--elen', required=False, type=str, metavar='<file>',
help='length model for exons [%(default)s]')
parser.add_argument('--ilen', required=False, type=str, metavar='<file>',
help='length model for introns [%(default)s]')
parser.add_argument('--introns', required=False, type=str, metavar='<file>',
help='use introns in gff for source of splice sites [%(default)s]')
parser.add_argument('--limit', required=False, type=int, default=20,
metavar='<int>', help='limit number of transcripts [%(default)i]')
parser.add_argument('--wdpwm', required=False, type=float, default=1.0,
metavar='<float>', help='dpwm weight [%(default).2f]')
parser.add_argument('--wapwm', required=False, type=float, default=1.0,
metavar='<float>', help='apwm weight [%(default).2f]')
parser.add_argument('--wemm', required=False, type=float, default=1.0,
metavar='<float>', help='emm weight [%(default).2f]')
parser.add_argument('--wimm', required=False, type=float, default=1.0,
metavar='<float>', help='imm weight [%(default).2f]')
parser.add_argument('--welen', required=False, type=float, default=1.0,
metavar='<float>', help='elen weight [%(default).2f]')
parser.add_argument('--wilen', required=False, type=float, default=1.0,
metavar='<float>', help='ilen weight [%(default).2f]')
parser.add_argument('--icost', required=False, type=float, default=0.0,
metavar='<float>', help='cost for each intron [%(default).2f]')
arg = parser.parse_args()
assert(arg.icost >= 0)
dpwm = isoform.read_pwm(arg.dpwm) if arg.dpwm else None
apwm = isoform.read_pwm(arg.apwm) if arg.apwm else None
elen = isoform.read_len(arg.elen) if arg.elen else None
ilen = isoform.read_len(arg.ilen) if arg.ilen else None
emm = isoform.read_markov(arg.emm) if arg.emm else None
imm = isoform.read_markov(arg.imm) if arg.imm else None
# generate isoforms
name, seq = next(isoform.read_fasta(arg.fasta))
txs, info = isoform.all_possible(seq, arg.min_intron, arg.min_exon,
arg.max_splice, arg.flank, gff=arg.introns)
# score isoforms
for tx in txs:
score = 0
if apwm: score += isoform.score_apwm(apwm, tx) * arg.wapwm
if dpwm: score += isoform.score_dpwm(dpwm, tx) * arg.wdpwm
if elen: score += isoform.score_elen(elen, tx) * arg.welen
if ilen: score += isoform.score_ilen(ilen, tx) * arg.wilen
if emm: score += isoform.score_emm(emm, tx) * arg.wemm
if imm: score += isoform.score_imm(imm, tx, dpwm, apwm) * arg.wimm
score -= len(tx['introns']) * arg.icost
tx['score'] = score
# summary output
print('# name:', name)
print('# length:', len(seq))
print('# donors:', info['donors'])
print('# acceptors:', info['acceptors'])
print('# trials:', info['trials'])
print('# isoforms:', len(txs))
print(f'# complexity: {isoform.complexity(txs):.4f}')
if arg.limit:
limit = arg.limit
if limit > len(txs): limit = len(txs)
else:
limit = len(txs)
txs = sorted(txs, key=lambda item: item['score'], reverse=True)
# calculate probability of each isoform
weight = []
total = 0
for i in range(limit):
tx = txs[i]
w = 2 ** tx['score']
weight.append(w)
total += w
prob = []
for w in weight: prob.append(w / total)
# create gff
chrom = name.split()[0]
src = 'asg'
cs = f'{chrom}\t{src}\t'
b = txs[0]['beg'] + 1
e = txs[0]['end'] + 1
gene = f'Gene-{chrom}'
print(f'{cs}gene\t{b}\t{e}\t.\t+\t.\tID={gene}\n')
for i in range(limit):
tx = txs[i]
b = tx['beg'] + 1
e = tx['end'] + 1
s = prob[i]
tid = f'tx-{chrom}-{i+1}'
print(f'{cs}mRNA\t{b}\t{e}\t{s:.4g}\t+\t.\tID={tid};Parent={gene}')
for exon in tx['exons']:
b = exon[0] + 1
e = exon[1] + 1
print(f'{cs}exon\t{b}\t{e}\t{s:.4g}\t+\t.\tParent={tid}')
for intron in tx['introns']:
b = intron[0] + 1
e = intron[1] + 1
print(f'{cs}intron\t{b}\t{e}\t{s:.4g}\t+\t.\tParent={tid}')
print()