This repository has been archived by the owner on Jul 1, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 1
/
alifoldz_analysis.py
149 lines (119 loc) · 5.39 KB
/
alifoldz_analysis.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
136
137
138
139
140
141
142
143
144
145
146
147
148
149
#!/usr/bin/env python
# coding: utf-8
from sarscov2_util import *
import os
from Bio import AlignIO
from scipy import stats
def make_stockholm_alignment(window_start, window_end, sequences, alignment_file):
# Make fasta file alignment with desired range
f = open('alignments/tmp_alignment.fa', 'w')
for sequence in sequences:
f.write('> %s\n' % sequence.name)
cur_seq = sequence.seq[window_start:window_end]
ii = 0
while ii < len(cur_seq):
f.write("%s\n" % cur_seq[ii:(ii+60)])
ii += 60
f.close()
# Make stockholm alignment file with desired range
align = AlignIO.read('alignments/tmp_alignment.fa', 'fasta')
AlignIO.write(align, alignment_file, 'stockholm')
os.remove('alignments/tmp_alignment.fa')
def make_fa_alignment(window_start, window_end, sequences, alignment_file):
# Make fasta file alignment with desired range
f = open(alignment_file, 'w')
for sequence in sequences:
f.write('> %s\n' % sequence.name)
cur_seq = sequence.seq[window_start:window_end]
ii = 0
while ii < len(cur_seq):
f.write("%s\n" % cur_seq[ii:(ii+60)])
ii += 60
f.close()
if __name__=='__main__':
sequences = get_sequences('alignments/betacorona-genome-ref-oc43.fa')
(full_ref_seq, ref_seq) = get_ref_seq(sequences)
# Make windowed alignments for use with rscape
ii = 0
window_id = 0
while ii < len(full_ref_seq):
window_start = ii
window_end = ii + 120
alignment_file = 'rscape/windows_betacov/alignment_' + str(window_id) + '.sto'
make_stockholm_alignment(window_start, window_end, sequences, alignment_file)
ii += 40
window_id += 1
sequences = get_sequences('alignments/RR_nCov_alignment2_021120_muscle.fa')
(full_ref_seq, ref_seq) = get_ref_seq(sequences)
# Make windowed alignments for use with alifoldz
ii = 0
window_id = 0
window_id_dict = {}
while ii < len(full_ref_seq):
window_start = ii
window_end = ii + 120
alignment_file = 'alifoldz/windows/alignment_' + str(window_id) + '.fa'
make_fa_alignment(window_start, window_end, sequences, alignment_file)
window_id_dict[window_id] = (window_start, window_end)
ii += 40
window_id += 1
# Determine alifoldz cutoff
# Compare alifoldz windows to rnaz windows:
f = open("alifoldz/alifoldz_shuffle_data.dat")
alifoldz_lines = f.readlines()
f.close()
# Chosen arbitrarily for now, rechoose based on shuffled sequences.
alifoldz_vals = []
for alifoldz_line in alifoldz_lines:
alifoldz_items = alifoldz_line.strip('\n').split(' ')
if alifoldz_items[3] == '':
continue
alifoldz_vals += [float(alifoldz_items[3])]
alifoldz_vals = np.array(alifoldz_vals)
CUTOFF = np.quantile(alifoldz_vals, 0.01)
print(CUTOFF)
aln_file_1 = 'alignments/RR_nCov_alignment2_021120_muscle.fa'
rnaz_windows_f = 'rnaz_data/rnaz_RR_nCov_alignment2_021120_muscle.out'
rnaz_windows_dict = get_rnaz_windows(rnaz_windows_f, aln_file_1)
# Compare alifoldz windows to rnaz windows:
f = open("alifoldz/alifoldz_data.dat")
alifoldz_lines = f.readlines()
f.close()
# Do alifoldz intervals with z scores below cutoff based on shuffled alifoldz
# overlap with RNAz intervals?
alifoldz_intervals = []
rnaz_overlaps = 0
num_cutoff = 0
f = open("results/rnaz_alifoldz_windows.csv", 'w')
for ii, alifoldz_line in enumerate(alifoldz_lines):
alifoldz_items = alifoldz_line.strip('\n').split(' ')
if alifoldz_items[3] == '':
continue
if float(alifoldz_items[3]) <= CUTOFF:
num_cutoff += 1
(int_start, int_end) = window_id_dict[ii]
ref_int_start = get_position(full_ref_seq, int_start)
ref_int_end = get_position(full_ref_seq, int_end) - 1
alifoldz_intervals += [(ref_int_start, ref_int_end)]
if (ref_int_start, ref_int_end) in rnaz_windows_dict.keys():
rnaz_overlaps += 1
(seq, rnaz_secstruct, rnaz_z, rnaz_p) = rnaz_windows_dict[(ref_int_start, ref_int_end)]
alifoldz_z = float(alifoldz_items[3])
f.write('%d,%d,%s,%s,%.2f,%.4f,%.2f\n' % (ref_int_start, ref_int_end, seq, rnaz_secstruct, rnaz_z, rnaz_p, alifoldz_z))
f.close()
total_rnaz = len(rnaz_windows_dict.keys())
total_alifoldz = num_cutoff
total_windows = len(window_id_dict.keys())
total_overlap = rnaz_overlaps
print("Total overlapping intervals: %d; Total RNAz intervals: %d; Total alifoldz intervals: %d; Total Windows: %d\n" % (total_overlap, total_rnaz, total_alifoldz, total_windows))
rv = stats.hypergeom(total_windows, total_alifoldz, total_rnaz)
p_value = 1 - rv.cdf(total_overlap)
print("P value for overlap: %.2E" % p_value)
# Do alifoldz intervals overlap with structured elements?
print("Structured elements that overlap with alifoldz windows:")
for region_key in regions.keys():
region_interval = regions[region_key]
region_start = get_position_full(full_ref_seq, region_interval[0])
region_end = get_position_full(full_ref_seq, region_interval[1])
if get_interval_overlap_single([region_start, region_end], alifoldz_intervals):
print(region_key)