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
|
#(c) 2013-2014 by Authors
#This file is a part of Ragout program.
#Released under the BSD license (see LICENSE file)
"""
Some helping function for alignment manipulations
"""
from __future__ import absolute_import
from collections import namedtuple, defaultdict
from six.moves import filter
AlignmentColumn = namedtuple("AlignmentColumn", ["ref", "qry"])
AlignmentRow = namedtuple("AlignmentRow", ["start", "end", "strand",
"seq_len", "seq_id"])
def group_by_chr(alignment):
by_chr = defaultdict(list)
for entry in alignment:
by_chr[entry.ref.seq_id].append(entry)
for chr_id in by_chr:
by_chr[chr_id].sort(key=lambda e: e.ref.start)
return by_chr
def join_collinear(alignment):
new_entries = []
def append_entry(start_entry, last_entry):
ref_row = AlignmentRow(start_entry.ref.start, last_entry.ref.end,
start_entry.ref.strand, start_entry.ref.seq_len,
start_entry.ref.seq_id)
qry_left, qry_right = sorted([start_entry.qry, last_entry.qry],
key=lambda r: r.start)
qry_row = AlignmentRow(qry_left.start, qry_right.end,
qry_left.strand, qry_left.seq_len,
qry_left.seq_id)
new_entries.append(AlignmentColumn(ref_row, qry_row))
by_chr = group_by_chr(alignment)
for chr_id in by_chr:
by_chr[chr_id].sort(key=lambda e: e.ref.start)
start_entry = None
last_entry = None
for entry in by_chr[chr_id]:
if not start_entry:
start_entry = entry
last_entry = entry
continue
#checking for connection consistency
ref_type = last_entry.ref.strand * entry.ref.strand
qry_left, qry_right = sorted([last_entry.qry, entry.qry],
key=lambda r: r.start)
qry_type = qry_left.strand * qry_right.strand
###
if (entry.qry.seq_id != last_entry.qry.seq_id or
ref_type != qry_type):
append_entry(start_entry, last_entry)
start_entry = entry
last_entry = entry
if start_entry:
append_entry(start_entry, last_entry)
return new_entries
def aln_len(row):
assert row.end >= row.start
return row.end - row.start
def filter_by_coverage(alignment, threshold):
by_name = defaultdict(list)
for entry in alignment:
by_name[entry.qry.seq_id].append(entry)
for name in by_name:
by_name[name].sort(key=lambda e: aln_len(e.qry), reverse=True)
len_filter = (lambda e: aln_len(e.qry) > threshold *
aln_len(by_name[name][0].qry))
by_name[name] = list(filter(len_filter, by_name[name]))
filtered_alignment = []
for ent_lst in by_name.values():
filtered_alignment.extend(ent_lst)
return filtered_alignment
class Hit:
def __init__(self, index, chr, coord, sign):
self.index = index
self.chr = chr
self.coord = coord
self.sign = sign
def __str__(self):
return ("+" if self.sign > 0 else "-") + str(self.index) + " : " + str(self.chr)
def get_order(alignment):
chr_len = defaultdict(int)
contig_len = defaultdict(int)
for entry in alignment:
contig_len[entry.qry.seq_id] = max(entry.qry.end,
contig_len[entry.qry.seq_id])
chr_len[entry.ref.seq_id] = max(entry.ref.end,
chr_len[entry.ref.seq_id])
by_chr = group_by_chr(alignment)
entry_ord = defaultdict(list)
contig_pos = 1
prev_start = None
for chr_id, alignment in by_chr.items():
for e in alignment:
if prev_start is not None and e.ref.start > prev_start:
contig_pos += 1
prev_start = e.ref.start
sign = e.ref.strand * e.qry.strand
entry_ord[e.qry.seq_id].append(Hit(contig_pos, chr_id,
e.ref.start, sign))
return entry_ord, chr_len, contig_len
|
