-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcontactprobability.py
176 lines (151 loc) · 5.49 KB
/
contactprobability.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
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
#!/usr/bin/env python
import numpy
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import subprocess
import sys
from bisect import bisect_left, bisect_right
from scipy import stats
import random
"""
1st argument is name of experiment
2nd argument is location of file containing contacts
3rd argument is genome (hg19 or mm10)
contacts should be prefiltered using awk
example with medium format merged_nodups:
<readname> <str1> <chr1> <pos1> <frag1> <str2> <chr2> <pos2> <frag2> <mapq1> <mapq2>
awk '{if($3 == $7 && $10 > 29 && $11 > 29){print $2"\t"$3"\t"$4"\t"$6"\t"$7"\t"$8}}' merged_nodups.txt > filtered_merged_nodups.txt
example with long format merged_nodups:
<str1> <chr1> <pos1> <frag1> <str2> <chr2> <pos2> <frag2> <mapq1> <cigar1> <sequence1> <mapq2> <cigar2> <sequence2> <readname1> <readname2>
awk '{if($2 == $6 && $9 > 29 && $12 > 29){print $1"\t"$2"\t"$3"\t"$5"\t"$6"\t"$7}}' merged_nodups.txt > filtered_merged_nodups.txt
"""
description = sys.argv[1]
file_location = sys.argv[2]
genome = str(sys.argv[3])
# chrom.sizes
chrom_dot_sizes = [249904550,243199373,198022430,191535534,180915260,\
171115067,159321559,146440111,141696573,135534747,\
135046619,133851895,115169878,107349540,102531392,\
90354753,81529607,78081510,59380841,63025520,48157577,\
51304566,155270560,59373566]
if genome == "mm10":
chrom_dot_sizes = [195471971,182113224,160039680,156508116,151834684,\
149736546,145441459,129401213,124595110,130694993,\
122082543,120129022,120421639,124902244,104043685,\
98207768,94987271,90702639,61431566,171031299]
def get_chrom_length(number):
if number=="X":
number=23
if genome == "mm10":
number=20
return chrom_dot_sizes[int(number)-1]
def totalpairs(chrlength,d1,d2):
if d2<=d1: poss=0
else: poss = (d2-d1-1)*chrlength-(d2*(d2-1)-d1*(d1+1))/2
return poss
def run_code(exp_name, hicfile):
figname=exp_name+'_genome_wide_cp.png'
figname_split=exp_name+'_iolr_genome_wide_cp.png'
histname=exp_name+'_hist_genome_wide_cp.txt'
actualhist=numpy.zeros(1000)
actualhist_split={}
for type in ['inner','outer','left','right']:
actualhist_split[type]=numpy.zeros(1000)
possiblehist=numpy.zeros(1000)
bin_edges=numpy.logspace(0.0,(numpy.log(get_chrom_length(1))/numpy.log(10)),1001)
# compute actual contacts
f=open(hicfile, 'r')
next=f.readline()
while next!="":
s1,chr1,x1,s2,chr2,x2=next.split()[:6]
chr1 = chr1.replace("chr","")
chr2 = chr2.replace("chr","")
if chr1 == chr2 and chr1 != "MT" and chr1 != "M" and chr1 != "Y":
actualhist[bisect_left(bin_edges,abs(int(x2)-int(x1)))-1]+=1
# check to make sure x2 corresponds to a further position
if int(x2)<int(x1):
s1, s2 = s2, s1
if s1=='0' and s2=='0':
type='right'
elif s1=='16' and s2=='16':
type='left'
elif s1=='0' and s2=='16':
type='inner'
elif s1=='16' and s2=='0':
type='outer'
actualhist_split[type][bisect_left(bin_edges,abs(int(x2)-int(x1)))-1]+=1
next=f.readline()
f.close()
# compute possible contacts
num_max_chroms = len(chrom_dot_sizes)
for i in range(1,num_max_chroms):
chr_num=str(i)
if i==num_max_chroms:
chr_num='X'
chrom_length=get_chrom_length(chr_num)
for j in range(1000):
possiblehist[j]+=totalpairs(chrom_length,int(bin_edges[j]),int(bin_edges[j+1])+1)
#g=open('poss_contacts.txt','w')
#for i in range(1000):
# g.write(str(possiblehist[i])+'\t'+str(bin_edges[i])+'\n')
#g.close()
# compute contact probability to make the final histogram
histfile=open(histname,"w")
for i in range(1000):
actual=actualhist[i]
if possiblehist[i]==0:
actualhist[i]=0
else:
actualhist[i]=actualhist[i]/float(possiblehist[i])
histfile.write(str(actualhist[i])+' '+str(bin_edges[i])+' '+str(actual)+' '+str(possiblehist[i])+'\n')
histfile.close()
###
# Plot histogram
###
plotxleft = 100
plotxright= 100000000
fit1left = 30000
fit1right = 300000
fit2left = 300000
fit2right = 3000000
plt.loglog(bin_edges[0:1000],actualhist)
plt.axis([plotxleft,plotxright,actualhist[-1]/100,actualhist[bisect_left(bin_edges,plotxleft)]*100])
plt.xlabel('Distance(bp)')
plt.ylabel('Contact Probability')
if plotxright > fit1right:
x1=bin_edges[bisect_left(bin_edges,fit1left):bisect_right(bin_edges,fit1right)]
x1c=actualhist[bisect_left(bin_edges,fit1left):bisect_right(bin_edges,fit1right)]
scal1=numpy.polyfit(numpy.log(x1),numpy.log(x1c),1)
y1=numpy.exp(scal1[0]*numpy.log(x1)+scal1[1]+.5)
r1=stats.pearsonr(x1c,y1)
plt.loglog(x1,y1,'r',linewidth=2.0)
plt.legend((exp_name,str(scal1[0])[:8]+" "+str(r1[0])[:8]))
else:
plt.legend((exp_name))
if plotxright > fit2right:
x2=bin_edges[bisect_left(bin_edges,fit2left):bisect_right(bin_edges,fit2right)]
x2c=actualhist[bisect_left(bin_edges,fit2left):bisect_right(bin_edges,fit2right)]
scal2=numpy.polyfit(numpy.log(x2),numpy.log(x2c),1)
y2=numpy.exp(scal2[0]*numpy.log(x2)+scal2[1]+.5)
r2=stats.pearsonr(x2c,y2)
plt.loglog(x2,y2,'g',linewidth=2.0)
plt.legend((exp_name, str(scal1[0])[:8]+" "+str(r1[0])[:8], str(scal2[0])[:8]+" "+str(r2[0])[:8]))
plt.savefig(figname)
plt.close('all')
# make plot for left/right/inner/outer-CP
for type,hist in actualhist_split.items():
for i in range(1000):
if possiblehist[i]==0:
hist[i]=0
else:
hist[i]=hist[i]/float(possiblehist[i])
plt.loglog(bin_edges[0:1000],hist,label=type)
plt.axis([plotxleft,plotxright,10e-10,10e-3])
plt.xlabel('Distance(bp)')
plt.ylabel('Contact Probability')
plt.title(exp_name)
plt.legend()
plt.savefig(figname_split)
if __name__=="__main__":
run_code(description, file_location)