PageRenderTime 26ms CodeModel.GetById 18ms app.highlight 5ms RepoModel.GetById 1ms app.codeStats 0ms

/scripts/microbes/create_nib_seq_loc_file.py

https://bitbucket.org/cistrome/cistrome-harvard/
Python | 88 lines | 77 code | 9 blank | 2 comment | 7 complexity | d3f936bf520e913e62906f1505d019ac MD5 | raw file
 1#!/usr/bin/env python
 2#Dan Blankenberg
 3
 4import sys, os
 5
 6assert sys.version_info[:2] >= ( 2, 4 )
 7
 8def __main__():
 9    base_dir = os.path.join( os.getcwd(), "bacteria" )
10    try:
11        base_dir = sys.argv[1]
12    except:
13        print "using default base_dir:", base_dir
14
15    loc_out = os.path.join( base_dir, "seq.loc" )
16    try:
17        loc_out = os.path.join( base_dir, sys.argv[2] )
18    except:
19        print "using default seq.loc:", loc_out
20
21
22
23    organisms = {}
24
25    loc_out = open( loc_out, 'wb' )
26
27    for result in os.walk( base_dir ):
28        this_base_dir, sub_dirs, files = result
29        for file in files:
30            if file[-5:] == ".info":
31                dict = {}
32                info_file = open( os.path.join( this_base_dir, file ), 'r' )
33                info = info_file.readlines()
34                info_file.close()
35                for line in info:
36                    fields = line.replace( "\n", "" ).split( "=" )
37                    dict[fields[0]]="=".join(fields[1:])
38                if 'genome project id' in dict.keys():
39                    name = dict['genome project id']
40                    if 'build' in dict.keys():
41                        name = dict['build']
42                    if name not in organisms.keys():
43                        organisms[name] = {'chrs':{}, 'base_dir':this_base_dir}
44                    for key in dict.keys():
45                        organisms[name][key] = dict[key]
46                else:
47                    if dict['organism'] not in organisms.keys():
48                        organisms[dict['organism']] = {'chrs':{},'base_dir':this_base_dir}
49                    organisms[dict['organism']]['chrs'][dict['chromosome']] = dict
50
51
52    for org in organisms:
53        org = organisms[org]
54        try:
55            build = org['genome project id']
56        except: continue
57        if 'build' in org:
58            build = org['build']
59        
60        seq_path = os.path.join( org['base_dir'], "seq" )
61        
62        #create seq dir, if exists go to next org
63        ##TODO: add better checking, i.e. for updating
64        try:
65            os.mkdir( seq_path )
66        except:
67            print "Skipping", build
68            #continue
69        
70        loc_out.write( "seq %s %s\n" % ( build, seq_path ) )
71        
72        #print org info
73        
74        for chr in org['chrs']:
75            chr = org['chrs'][chr]
76            
77            fasta_file = os.path.join( org['base_dir'], "%s.fna" % chr['chromosome'] )
78            nib_out_file = os.path.join( seq_path, "%s.nib "% chr['chromosome'] )
79            #create nibs using faToNib binary
80            #TODO: when bx supports writing nib, use it here instead
81            command = "faToNib %s %s" % ( fasta_file, nib_out_file )
82            os.system( command )
83    
84    loc_out.close()
85    
86    
87
88if __name__ == "__main__": __main__()