|
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 FILE_TYPE_TO_TRACK_TYPE = { 'bed': None, 'bedstrict': None, 'bed6': None, 'bed12': None, 'bedgraph':'bedGraph', 'wig':'wiggle_0' }
|
|
|
15 CHUNK_SIZE = 2**20 #1mb
|
|
|
16
|
|
|
17 def get_track_line_is_interval( file_type, name, description, color, visibility ):
|
|
|
18 if file_type in FILE_TYPE_TO_TRACK_TYPE:
|
|
|
19 track_type = FILE_TYPE_TO_TRACK_TYPE[ file_type ]
|
|
|
20 is_interval = False
|
|
|
21 else:
|
|
|
22 track_type = None
|
|
|
23 is_interval = True
|
|
|
24 track_line = 'track '
|
|
|
25 if track_type:
|
|
|
26 track_line += 'type=%s ' % ( track_type )
|
|
|
27 track_line += 'name="%s" description="%s" color=%s visibility=%s\n' % ( name, description, color, visibility )
|
|
|
28 return track_line, is_interval
|
|
|
29
|
|
|
30 args = sys.argv[1:]
|
|
|
31
|
|
|
32 out_fname = args.pop(0)
|
|
|
33 out = open( out_fname, "w" )
|
|
|
34
|
|
|
35 num_tracks = 0
|
|
|
36 skipped_lines = 0
|
|
|
37 first_invalid_line = 0
|
|
|
38 while args:
|
|
|
39 # Suck in one dataset worth of arguments
|
|
|
40 in_fname = args.pop(0)
|
|
|
41 file_type = args.pop(0)
|
|
|
42 colspec = args.pop(0)
|
|
|
43 name = args.pop(0)
|
|
|
44 description = args.pop(0)
|
|
|
45 color = args.pop(0).replace( '-', ',' )
|
|
|
46 visibility = args.pop(0)
|
|
|
47 track_line, is_interval = get_track_line_is_interval( file_type, name, description, color, visibility )
|
|
|
48 # Do the work
|
|
|
49 in_file = open( in_fname )
|
|
|
50 out.write( track_line )
|
|
|
51 if not is_interval:
|
|
|
52 while True:
|
|
|
53 chunk = in_file.read( CHUNK_SIZE )
|
|
|
54 if chunk:
|
|
|
55 out.write( chunk )
|
|
|
56 else:
|
|
|
57 break
|
|
|
58 else:
|
|
|
59 # Assume type is interval (don't pass this script anything else!)
|
|
|
60 try:
|
|
|
61 c, s, e, st = [ int( x ) - 1 for x in colspec.split( "," ) ]
|
|
|
62 except:
|
|
|
63 try:
|
|
|
64 c, s, e = [ int( x ) - 1 for x in colspec.split( "," )[:3] ]
|
|
|
65 st = -1 #strand column is absent
|
|
|
66 except:
|
|
|
67 stop_err( "Columns in interval file invalid for UCSC custom track." )
|
|
|
68
|
|
|
69 i = 0
|
|
|
70 for i, line in enumerate( in_file ):
|
|
|
71 line = line.rstrip( '\r\n' )
|
|
|
72 if line and not line.startswith( '#' ):
|
|
|
73 fields = line.split( "\t" )
|
|
|
74 if st > 0:
|
|
|
75 #strand column is present
|
|
|
76 try:
|
|
|
77 out.write( "%s\t%s\t%s\t%d\t0\t%s\n" % ( fields[c], fields[s], fields[e], i, fields[st] ) )
|
|
|
78 except:
|
|
|
79 skipped_lines += 1
|
|
|
80 if not first_invalid_line:
|
|
|
81 first_invalid_line = i+1
|
|
|
82 else:
|
|
|
83 try:
|
|
|
84 out.write( "%s\t%s\t%s\n" % ( fields[c], fields[s], fields[e] ) )
|
|
|
85 except:
|
|
|
86 skipped_lines += 1
|
|
|
87 if not first_invalid_line:
|
|
|
88 first_invalid_line = i+1
|
|
|
89 out.write( "\n" ) #separating newline
|
|
|
90 num_tracks += 1
|
|
|
91
|
|
|
92 out.close()
|
|
|
93
|
|
|
94 print "Generated a custom track containing %d subtracks." % num_tracks
|
|
|
95 if skipped_lines:
|
|
|
96 print "Skipped %d invalid lines starting at #%d" % ( skipped_lines, first_invalid_line )
|
|
|
97
|
|
|
98
|
|
|
99
|
