2
|
1 #!/usr/bin/python
|
|
2 # script for computing overlap signatures from a bowtie output
|
|
3 # Christophe Antoniewski <drosofff@gmail.com>
|
|
4 # Usage piRNAsignature.py <1:input> <2:format of input> <3:minsize query> <4:maxsize query> <5:minsize target> <6:maxsize target>
|
|
5 # <7:minscope> <8:maxscope> <9:output> <10:bowtie index> <11:procedure option> <12: graph (global or lattice)>
|
|
6 # <13: R code>
|
|
7
|
|
8 import sys, subprocess
|
|
9 from smRtools import *
|
|
10 from collections import defaultdict # test whether it is required
|
|
11
|
|
12 if sys.argv[11] == "--extract_index":
|
|
13 if sys.argv[2] == "tabular":
|
|
14 Genome = HandleSmRNAwindows (sys.argv[1],"tabular",sys.argv[10],"bowtieIndex")
|
|
15 elif sys.argv[2] == "sam":
|
|
16 Genome = HandleSmRNAwindows (sys.argv[1],"sam",sys.argv[10],"bowtieIndex")
|
|
17 else:
|
|
18 Genome = HandleSmRNAwindows (sys.argv[1],"bam",sys.argv[10],"bowtieIndex")
|
|
19 else:
|
|
20 if sys.argv[2] == "tabular":
|
|
21 Genome = HandleSmRNAwindows (sys.argv[1],"tabular",sys.argv[10],"fastaSource")
|
|
22 elif sys.argv[2] == "sam":
|
|
23 Genome = HandleSmRNAwindows (sys.argv[1],"sam",sys.argv[10],"fastaSource")
|
|
24 else:
|
|
25 Genome = HandleSmRNAwindows (sys.argv[1],"bam",sys.argv[10],"fastaSource")
|
|
26 # this decisional tree may be simplified if sam and bam inputs are treated the same way by pysam
|
|
27
|
|
28 # replace objDic by Genome.instanceDict or... objDic = Genome.instanceDict
|
|
29 objDic = Genome.instanceDict
|
|
30
|
|
31 minquery = int(sys.argv[3])
|
|
32 maxquery = int(sys.argv[4])
|
|
33 mintarget = int(sys.argv[5])
|
|
34 maxtarget = int(sys.argv[6])
|
|
35 minscope = int(sys.argv[7])
|
|
36 maxscope = int(sys.argv[8]) + 1
|
|
37 general_frequency_table = dict ([(i,0) for i in range(minscope,maxscope)])
|
|
38 general_percent_table = dict ([(i,0) for i in range(minscope,maxscope)])
|
|
39 OUT = open (sys.argv[9], "w")
|
|
40
|
|
41 if sys.argv[12] == "global":
|
|
42 ###### for normalized summing of local_percent_table(s)
|
|
43 readcount_dic = {}
|
|
44 Total_read_in_objDic = 0
|
|
45 for item in objDic:
|
|
46 readcount_dic[item] = objDic[item].readcount(minquery, maxquery)
|
|
47 Total_read_in_objDic += readcount_dic[item]
|
|
48 ######
|
|
49 for x in (objDic):
|
|
50 local_frequency_table = objDic[x].signature( minquery, maxquery, mintarget, maxtarget, range(minscope,maxscope) )
|
|
51 local_percent_table = objDic[x].hannon_signature( minquery, maxquery, mintarget, maxtarget, range(minscope,maxscope) )
|
|
52 try:
|
|
53 for overlap in local_frequency_table.keys():
|
|
54 general_frequency_table[overlap] = general_frequency_table.get(overlap, 0) + local_frequency_table[overlap]
|
|
55 except:
|
|
56 pass
|
|
57 try:
|
|
58 for overlap in local_percent_table.keys():
|
|
59 general_percent_table[overlap] = general_percent_table.get(overlap, 0) + (1./Total_read_in_objDic*readcount_dic[x]*local_percent_table[overlap])
|
|
60 except:
|
|
61 pass
|
|
62 print >> OUT, "overlap\tnum of pairs\tprobability"
|
|
63 for classe in sorted(general_frequency_table):
|
|
64 print >> OUT, "%i\t%i\t%f" % (classe, general_frequency_table[classe], general_percent_table[classe])
|
|
65
|
|
66 else:
|
|
67 print >> OUT, "overlap\tnum of pairs\tprobability\titem"
|
|
68 for x in (objDic):
|
|
69 local_frequency_table = objDic[x].signature( minquery, maxquery, mintarget, maxtarget, range(minscope,maxscope) )
|
|
70 local_percent_table = objDic[x].hannon_signature( minquery, maxquery, mintarget, maxtarget, range(minscope,maxscope) )
|
|
71 for classe in range(minscope,maxscope):
|
|
72 print >> OUT, "%i\t%i\t%f\t%s" % (classe, local_frequency_table[classe], local_percent_table[classe], x)
|
|
73
|
|
74 OUT.close()
|
|
75
|
|
76 ## Run the R script that is defined in the xml using the Rscript binary provided with R.
|
|
77 R_command="Rscript "+ sys.argv[13]
|
|
78 process = subprocess.Popen(R_command.split())
|