0
|
1 #!/usr/bin/env python
|
|
2 #Dan Blankenberg
|
|
3 """
|
|
4 Reads a list of intervals and a maf. Outputs a new set of intervals with statistics appended.
|
|
5 """
|
|
6
|
|
7 import sys
|
|
8 from galaxy import eggs
|
|
9 import pkg_resources; pkg_resources.require( "bx-python" )
|
|
10 import bx.intervals.io
|
|
11 from bx.bitset import BitSet
|
|
12 from galaxy.tools.util import maf_utilities
|
|
13
|
|
14 assert sys.version_info[:2] >= ( 2, 4 )
|
|
15
|
|
16 def __main__():
|
|
17 maf_source_type = sys.argv.pop( 1 )
|
|
18 input_maf_filename = sys.argv[1].strip()
|
|
19 input_interval_filename = sys.argv[2].strip()
|
|
20 output_filename = sys.argv[3].strip()
|
|
21 dbkey = sys.argv[4].strip()
|
|
22 try:
|
|
23 chr_col = int( sys.argv[5].strip() ) - 1
|
|
24 start_col = int( sys.argv[6].strip() ) - 1
|
|
25 end_col = int( sys.argv[7].strip() ) - 1
|
|
26 except:
|
|
27 print >>sys.stderr, "You appear to be missing metadata. You can specify your metadata by clicking on the pencil icon associated with your interval file."
|
|
28 sys.exit()
|
|
29 summary = sys.argv[8].strip()
|
|
30 if summary.lower() == "true": summary = True
|
|
31 else: summary = False
|
|
32
|
|
33 mafIndexFile = "%s/maf_index.loc" % sys.argv[9]
|
|
34 try:
|
|
35 maf_index_filename = sys.argv[10].strip()
|
|
36 except:
|
|
37 maf_index_filename = None
|
|
38 index = index_filename = None
|
|
39 if maf_source_type == "user":
|
|
40 #index maf for use here
|
|
41 index, index_filename = maf_utilities.open_or_build_maf_index( input_maf_filename, maf_index_filename, species = [dbkey] )
|
|
42 if index is None:
|
|
43 print >>sys.stderr, "Your MAF file appears to be malformed."
|
|
44 sys.exit()
|
|
45 elif maf_source_type == "cached":
|
|
46 #access existing indexes
|
|
47 index = maf_utilities.maf_index_by_uid( input_maf_filename, mafIndexFile )
|
|
48 if index is None:
|
|
49 print >> sys.stderr, "The MAF source specified (%s) appears to be invalid." % ( input_maf_filename )
|
|
50 sys.exit()
|
|
51 else:
|
|
52 print >>sys.stdout, 'Invalid source type specified: %s' % maf_source_type
|
|
53 sys.exit()
|
|
54
|
|
55 out = open(output_filename, 'w')
|
|
56
|
|
57 num_region = None
|
|
58 species_summary = {}
|
|
59 total_length = 0
|
|
60 #loop through interval file
|
|
61 for num_region, region in enumerate( bx.intervals.io.NiceReaderWrapper( open( input_interval_filename, 'r' ), chrom_col = chr_col, start_col = start_col, end_col = end_col, fix_strand = True, return_header = False, return_comments = False ) ):
|
|
62 src = "%s.%s" % ( dbkey, region.chrom )
|
|
63 region_length = region.end - region.start
|
|
64 total_length += region_length
|
|
65 coverage = { dbkey: BitSet( region_length ) }
|
|
66
|
|
67
|
|
68 for block in index.get_as_iterator( src, region.start, region.end ):
|
|
69 for spec in maf_utilities.get_species_in_block( block ):
|
|
70 if spec not in coverage: coverage[spec] = BitSet( region_length )
|
|
71 for block in maf_utilities.iter_blocks_split_by_species( block ):
|
|
72 if maf_utilities.component_overlaps_region( block.get_component_by_src( src ), region ):
|
|
73 #need to chop and orient the block
|
|
74 block = maf_utilities.orient_block_by_region( maf_utilities.chop_block_by_region( block, src, region ), src, region, force_strand = '+' )
|
|
75 start_offset, alignment = maf_utilities.reduce_block_by_primary_genome( block, dbkey, region.chrom, region.start )
|
|
76 for i in range( len( alignment[dbkey] ) ):
|
|
77 for spec, text in alignment.items():
|
|
78 if text[i] != '-':
|
|
79 coverage[spec].set( start_offset + i )
|
|
80 if summary:
|
|
81 #record summary
|
|
82 for key in coverage.keys():
|
|
83 if key not in species_summary: species_summary[key] = 0
|
|
84 species_summary[key] = species_summary[key] + coverage[key].count_range()
|
|
85 else:
|
|
86 #print coverage for interval
|
|
87 coverage_sum = coverage[dbkey].count_range()
|
|
88 out.write( "%s\t%s\t%s\t%s\n" % ( "\t".join( region.fields ), dbkey, coverage_sum, region_length - coverage_sum ) )
|
|
89 keys = coverage.keys()
|
|
90 keys.remove( dbkey )
|
|
91 keys.sort()
|
|
92 for key in keys:
|
|
93 coverage_sum = coverage[key].count_range()
|
|
94 out.write( "%s\t%s\t%s\t%s\n" % ( "\t".join( region.fields ), key, coverage_sum, region_length - coverage_sum ) )
|
|
95 if summary:
|
|
96 out.write( "#species\tnucleotides\tcoverage\n" )
|
|
97 for spec in species_summary:
|
|
98 out.write( "%s\t%s\t%.4f\n" % ( spec, species_summary[spec], float( species_summary[spec] ) / total_length ) )
|
|
99 out.close()
|
|
100 if num_region is not None:
|
|
101 print "%i regions were processed with a total length of %i." % ( num_region + 1, total_length )
|
|
102 maf_utilities.remove_temp_index_file( index_filename )
|
|
103
|
|
104 if __name__ == "__main__": __main__()
|