Mercurial > repos > artbio > blast_unmatched

comparison blast_unmatched.py @ 0:f3b63b59a1ea draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/ARTbio/tools-artbio/tree/master/tools/blast_unmatched commit 5bd801feb838592fbb1f6dd68b5f1a480042da40
author artbio
date Tue, 03 Oct 2017 07:19:17 -0400
parents
children 50c1fa95a076
comparison
equal deleted inserted replaced
-1:000000000000 0:f3b63b59a1ea
1 #!/usr/bin/env python3
2
3 import optparse
4
5
6 def parse_options():
7 """
8 Parse the options guiven to the script
9 """
10 parser = optparse.OptionParser(description='Get unmatched blast queries')
11 parser.add_option('-f','--fasta', dest='fasta_file', help='Query fasta file\
12 used during blast')
13 parser.add_option('-b','--blast', dest='blast_file', help='Blast tabular\
14 output (queries in 1rst column)')
15 parser.add_option('-o','--output', dest='output_file', help='Output file name')
16 (options, args) = parser.parse_args()
17 if len(args) > 0:
18 parser.error('Wrong number of arguments')
19 return options
20
21 def get_matched(blast_file):
22 """
23 Get a dictionary of all the queries that got a match
24 """
25 matched = dict()
26 blast_file_handle = open(blast_file, 'r')
27 for line in blast_file_handle.readlines():
28 fields = line.split("\t")
29 query_id = fields[0]
30 matched[query_id] = 1
31 blast_file_handle.close()
32 return matched
33
34 def get_unmatched(output_file, fasta_file, matched):
35 """
36 Compares matched queries to query fasta file and print unmatched to ouput
37 """
38 output_file_handle = open(output_file, 'w')
39 fasta_file_handle = open(fasta_file, 'r')
40 unmatched = False
41 for line in fasta_file_handle.readlines():
42 if line.startswith('>'):
43 subline = line[1:100].rstrip() #qid are 100chars long in blast
44 if subline not in matched:
45 output_file_handle.write(line)
46 unmatched = True
47 else:
48 unmatched = False
49 elif unmatched:
50 output_file_handle.write(line)
51 fasta_file_handle.close()
52 output_file_handle.close()
53
54 def __main__():
55 opts = parse_options()
56 matched = get_matched(opts.blast_file)
57 get_unmatched(opts.output_file, opts.fasta_file, matched)
58
59 if __main__():
60 __main__()