Mercurial > repos > vipints > fml_gff3togtf

comparison gff_to_bed.py @ 10:c42c69aa81f8

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
fixed manually the upload of version 2.1.0 - deleted accidentally added files to the repo
author vipints <vipin@cbio.mskcc.org>
date Thu, 23 Apr 2015 18:01:45 -0400
parents
children
comparison
equal deleted inserted replaced
9:7d67331368f3 10:c42c69aa81f8
1 #!/usr/bin/env python
2 """
3 Convert genome annotation data in GFF/GTF to a 12 column BED format.
4 BED format typically represents the transcript models.
5
6 Usage: python gff_to_bed.py in.gff > out.bed
7
8 Requirement:
9 GFFParser.py: https://github.com/vipints/GFFtools-GX/blob/master/GFFParser.py
10
11 Copyright (C)
12 2009-2012 Friedrich Miescher Laboratory of the Max Planck Society, Tubingen, Germany.
13 2012-2015 Memorial Sloan Kettering Cancer Center New York City, USA.
14 """
15
16 import re
17 import sys
18 import GFFParser
19
20 def limitBEDWrite(tinfo):
21 """
22 Write a three column BED file
23
24 @args tinfo: list of genes
25 @type tinfo: numpy object
26 """
27
28 for contig_id, feature in tinfo.items():
29 uns_line = dict()
30 for tid, tloc in feature.items():
31 uns_line[(int(tloc[0])-1, int(tloc[1]))]=1
32 for ele in sorted(uns_line):
33 pline = [contig_id,
34 str(ele[0]-1),
35 str(ele[1])]
36
37 sys.stdout.write('\t'.join(pline)+"\n")
38
39
40 def writeBED(tinfo):
41 """
42 writing result files in bed format
43
44 @args tinfo: list of genes
45 @type tinfo: numpy object
46 """
47
48 for ent1 in tinfo:
49 child_flag = False
50
51 for idx, tid in enumerate(ent1['transcripts']):
52 child_flag = True
53 exon_cnt = len(ent1['exons'][idx])
54 exon_len = ''
55 exon_cod = ''
56 rel_start = None
57 rel_stop = None
58 for idz, ex_cod in enumerate(ent1['exons'][idx]):#check for exons of corresponding transcript
59 exon_len += '%d,' % (ex_cod[1]-ex_cod[0]+1)
60 if idz == 0: #calculate the relative start position
61 exon_cod += '0,'
62 rel_start = int(ex_cod[0])-1
63 rel_stop = int(ex_cod[1])
64 else:
65 exon_cod += '%d,' % (ex_cod[0]-1-rel_start) ## shifting the coordinates to zero
66 rel_stop = int(ex_cod[1])
67
68 if exon_len:
69 score = 0
70 score = ent1['transcript_score'][idx] if ent1['transcript_score'].any() else score ## getting the transcript score
71 out_print = [ent1['chr'],
72 str(rel_start),
73 str(rel_stop),
74 tid[0],
75 str(score),
76 ent1['strand'],
77 str(rel_start),
78 str(rel_stop),
79 '0',
80 str(exon_cnt),
81 exon_len,
82 exon_cod]
83 sys.stdout.write('\t'.join(out_print)+"\n")
84
85 if not child_flag: # file just contains only a single parent type i.e, gff3 defines only one feature type
86 score = 0
87 score = ent1['transcript_score'][0] if ent1['transcript_score'].any() else score
88
89 out_print = [ent1['chr'],
90 '%d' % int(ent1['start'])-1,
91 '%d' % int(ent1['stop']),
92 ent1['name'],
93 str(score),
94 ent1['strand'],
95 '%d' % int(ent1['start']),
96 '%d' % int(ent1['stop']),
97 '0',
98 '1',
99 '%d,' % (int(ent1['stop'])-int(ent1['start'])+1),
100 '0,']
101
102 sys.stdout.write('\t'.join(out_print)+"\n")
103
104
105 def __main__():
106 try:
107 query_file = sys.argv[1]
108 except:
109 print __doc__
110 sys.exit(-1)
111
112 Transcriptdb = GFFParser.Parse(query_file)
113 writeBED(Transcriptdb)
114
115 if __name__ == "__main__":
116 __main__()