Mercurial > repos > blankenberg > naive_variant_caller
comparison tools/naive_variant_caller.py @ 1:ae6edc0012ba
Populate naive_variant_caller repository.
author | Daniel Blankenberg <dan@bx.psu.edu> |
---|---|
date | Thu, 29 Aug 2013 10:54:14 -0400 |
parents | |
children | 8398666758e3 |
comparison
equal
deleted
inserted
replaced
0:0fa83c466e9d | 1:ae6edc0012ba |
---|---|
1 #Dan Blankenberg | |
2 import sys | |
3 import optparse | |
4 | |
5 from pyBamParser.bam import Reader | |
6 from pyBamTools.genotyping.naive import VCFReadGroupGenotyper | |
7 | |
8 def main(): | |
9 #Parse Command Line | |
10 parser = optparse.OptionParser() | |
11 parser.add_option( '-b', '--bam', dest='bam_file', action='append', type="string", default=[], help='BAM filename, optionally index filename. Multiple allowed.' ) | |
12 parser.add_option( '-i', '--index', dest='index_file', action='append', type="string", default=[], help='optionally index filename. Multiple allowed.' ) | |
13 parser.add_option( '-o', '--output_vcf_filename', dest='output_vcf_filename', action='store', default = None, type="string", help='Output VCF filename' ) | |
14 parser.add_option( '-r', '--reference_genome_filename', dest='reference_genome_filename', action='store', default = None, type="string", help='Input reference file' ) | |
15 parser.add_option( '-v', '--variants_only', dest='variants_only', action='store_true', default = False, help='Report only sites with a possible variant allele.' ) | |
16 parser.add_option( '-s', '--use_strand', dest='use_strand', action='store_true', default = False, help='Report counts by strand' ) | |
17 parser.add_option( '-p', '--ploidy', dest='ploidy', action='store', type="int", default=2, help='Ploidy. Default=2.' ) | |
18 parser.add_option( '-d', '--min_support_depth', dest='min_support_depth', action='store', type="int", default=0, help='Minimum number of reads needed to consider a REF/ALT. Default=0.' ) | |
19 parser.add_option( '-q', '--min_base_quality', dest='min_base_quality', action='store', type="int", default=None, help='Minimum base quality.' ) | |
20 parser.add_option( '-m', '--min_mapping_quality', dest='min_mapping_quality', action='store', type="int", default=None, help='Minimum mapping.' ) | |
21 parser.add_option( '-t', '--coverage_dtype', dest='coverage_dtype', action='store', type="string", default='uint8', help='dtype to use for coverage array' ) | |
22 parser.add_option( '--allow_out_of_bounds_positions', dest='allow_out_of_bounds_positions', action='store_true', default = False, help='Allows out of bounds positions to not throw fatal errors' ) | |
23 parser.add_option( '--region', dest='region', action='append', type="string", default=[], help='region' ) | |
24 (options, args) = parser.parse_args() | |
25 | |
26 if len( options.bam_file ) == 0: | |
27 print >>sys.stderr, 'You must provide at least one bam (-b) file.' | |
28 parser.print_help( sys.stderr ) | |
29 sys.exit( 1 ) | |
30 if options.index_file: | |
31 assert len( options.index_file ) == len( options.bam_file ), "If you provide a name for an index file, you must provide the index name for all bam files." | |
32 bam_files = zip( options.bam_file, options.index_file ) | |
33 else: | |
34 bam_files = [ ( x, ) for x in options.bam_file ] | |
35 if not options.reference_genome_filename: | |
36 print >> sys.stderr, "Warning: Reference file has not been specified. Providing a reference genome is highly recommended." | |
37 if options.output_vcf_filename: | |
38 out = open( options.output_vcf_filename, 'wb' ) | |
39 else: | |
40 out = sys.stdout | |
41 | |
42 regions = [] | |
43 if options.region: | |
44 for region in options.region: | |
45 region_split = region.split( ":" ) | |
46 region = region_split.pop( 0 ) | |
47 if region_split: | |
48 region_split = filter( bool, region_split[0].split( '-' ) ) | |
49 if region_split: | |
50 if len( region_split ) != 2: | |
51 print >> sys.stderr, "You must specify both a start and an end, or only a chromosome when specifying regions." | |
52 cleanup_before_exit( tmp_dir ) | |
53 sys.exit( 1 ) | |
54 region = tuple( [ region ] + map( int, region_split ) ) | |
55 regions.append( region ) | |
56 | |
57 coverage = VCFReadGroupGenotyper( map( lambda x: Reader( *x ), bam_files ), options.reference_genome_filename, dtype=options.coverage_dtype, | |
58 min_support_depth=options.min_support_depth, min_base_quality=options.min_base_quality, min_mapping_quality=options.min_mapping_quality, | |
59 restrict_regions=regions, use_strand=options.use_strand, allow_out_of_bounds_positions=options.allow_out_of_bounds_positions ) | |
60 for line in coverage.iter_vcf( ploidy=options.ploidy, variants_only=options.variants_only ): | |
61 out.write( "%s\n" % line ) | |
62 out.close() | |
63 | |
64 if __name__ == "__main__": main() |