|
0
|
1 #!/usr/bin/env python
|
|
|
2 """
|
|
|
3 Build a UCSC genome browser custom track file
|
|
|
4 """
|
|
|
5
|
|
|
6 import sys, os
|
|
|
7
|
|
|
8 assert sys.version_info[:2] >= ( 2, 4 )
|
|
|
9
|
|
|
10 def stop_err( msg ):
|
|
|
11 sys.stderr.write( msg )
|
|
|
12 sys.exit()
|
|
|
13
|
|
|
14 args = sys.argv[1:]
|
|
|
15
|
|
|
16 out_fname = args.pop(0)
|
|
|
17 out = open( out_fname, "w" )
|
|
|
18
|
|
|
19 num_tracks = 0
|
|
|
20 skipped_lines = 0
|
|
|
21 first_invalid_line = 0
|
|
|
22 while args:
|
|
|
23 # Suck in one dataset worth of arguments
|
|
|
24 in_fname = args.pop(0)
|
|
|
25 type = args.pop(0)
|
|
|
26 colspec = args.pop(0)
|
|
|
27 name = args.pop(0)
|
|
|
28 description = args.pop(0)
|
|
|
29 color = args.pop(0).replace( '-', ',' )
|
|
|
30 visibility = args.pop(0)
|
|
|
31 # Do the work
|
|
|
32 if type == "wig":
|
|
|
33 print >> out, '''track type=wiggle_0 name="%s" description="%s" color=%s visibility=%s''' \
|
|
|
34 % ( name, description, color, visibility )
|
|
|
35 for i, line in enumerate( file( in_fname ) ):
|
|
|
36 print >> out, line,
|
|
|
37 print >> out
|
|
|
38 elif type == "bed":
|
|
|
39 print >> out, '''track name="%s" description="%s" color=%s visibility=%s''' \
|
|
|
40 % ( name, description, color, visibility )
|
|
|
41 for i, line in enumerate( file( in_fname ) ):
|
|
|
42 print >> out, line,
|
|
|
43 print >> out
|
|
|
44 else:
|
|
|
45 # Assume type is interval (don't pass this script anything else!)
|
|
|
46 try:
|
|
|
47 c, s, e, st = [ int( x ) - 1 for x in colspec.split( "," ) ]
|
|
|
48 except:
|
|
|
49 try:
|
|
|
50 c, s, e = [ int( x ) - 1 for x in colspec.split( "," )[:3] ]
|
|
|
51 st = -1 #strand column is absent
|
|
|
52 except:
|
|
|
53 stop_err( "Columns in interval file invalid for UCSC custom track." )
|
|
|
54
|
|
|
55 print >> out, '''track name="%s" description="%s" color=%s visibility=%s''' \
|
|
|
56 % ( name, description, color, visibility )
|
|
|
57 i = 0
|
|
|
58 for i, line in enumerate( file( in_fname ) ):
|
|
|
59 line = line.rstrip( '\r\n' )
|
|
|
60 if line and not line.startswith( '#' ):
|
|
|
61 fields = line.split( "\t" )
|
|
|
62 if st > 0:
|
|
|
63 #strand column is present
|
|
|
64 try:
|
|
|
65 print >> out, "%s\t%s\t%s\t%d\t0\t%s" % ( fields[c], fields[s], fields[e], i, fields[st] )
|
|
|
66 except:
|
|
|
67 skipped_lines += 1
|
|
|
68 if not first_invalid_line:
|
|
|
69 first_invalid_line = i+1
|
|
|
70 else:
|
|
|
71 try:
|
|
|
72 print >> out, "%s\t%s\t%s" % ( fields[c], fields[s], fields[e] )
|
|
|
73 except:
|
|
|
74 skipped_lines += 1
|
|
|
75 if not first_invalid_line:
|
|
|
76 first_invalid_line = i+1
|
|
|
77 print >> out
|
|
|
78 num_tracks += 1
|
|
|
79
|
|
|
80 out.close()
|
|
|
81
|
|
|
82 print "Generated a custom track containing %d subtracks." % num_tracks
|
|
|
83 if skipped_lines:
|
|
|
84 print "Skipped %d invalid lines starting at #%d" % ( skipped_lines, first_invalid_line )
|
|
|
85
|
|
|
86
|
|
|
87
|
