Mercurial > repos > jjohnson > rsem_datatypes

comparison rsem.py @ 0:77151afcd323

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author jjohnson
date Mon, 10 Mar 2014 08:09:01 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:77151afcd323
1 """
2 RSEM datatypes
3 """
4 import os,os.path,re,sys
5 from galaxy.datatypes.data import get_file_peek
6 from galaxy.datatypes.images import Html
7 from galaxy.datatypes.tabular import Tabular
8 from galaxy.datatypes.metadata import MetadataElement
9 from galaxy.datatypes.sniff import get_headers
10
11 import logging
12
13 log = logging.getLogger(__name__)
14
15 class RsemIsoformsResults( Tabular ):
16 file_ext = "rsem.isoforms.results"
17 """
18 required columns:
19 transcript_id gene_id length effective_length expected_count TPM FPKM IsoPct
20 optional columns:
21 pme_expected_count pme_TPM pme_FPKM IsoPct_from_pme_TPM TPM_ci_lower_bound TPM_ci_upper_bound FPKM_ci_lower_bound FPKM_ci_upper_bound
22 """
23 def __init__(self, **kwd):
24 Tabular.__init__( self, **kwd )
25 """Initialize RsemResults datatype"""
26 self.comment_lines = 1
27 def sniff( self, filename ):
28 headers = get_headers( filename, '\n', count=1 )
29 return len(headers) > 0 and len(headers[0]) >= 8 and headers[0][0] == "transcript_id" and headers[0][1] == "gene_id" and headers[0][6] == "FPKM"
30 def set_meta( self, dataset, **kwd ):
31 Tabular.set_meta( self, dataset, skip=None, **kwd )
32
33 class RsemGenesResults( Tabular ):
34 file_ext = "rsem.genes.results"
35 """
36 required columns:
37 gene_id transcript_id(s) length effective_length expected_count TPM FPKM
38 optional columns:
39 pme_expected_count pme_TPM pme_FPKM TPM_ci_lower_bound TPM_ci_upper_bound FPKM_ci_lower_bound FPKM_ci_upper_bound
40 """
41 def __init__(self, **kwd):
42 Tabular.__init__( self, **kwd )
43 """Initialize RsemResults datatype"""
44 self.comment_lines = 1
45 def sniff( self, filename ):
46 headers = get_headers( filename, '\n', count=1 )
47 return len(headers) > 0 and len(headers[0]) >= 7 and headers[0][0] == "gene_id" and headers[0][1].startswith("transcript_id") and headers[0][6] == "FPKM"
48 def set_meta( self, dataset, **kwd ):
49 Tabular.set_meta( self, dataset, skip=None, **kwd )
50
51 class RsemReference( Html ):
52 """Class describing an RSEM reference"""
53 MetadataElement( name='reference_name', default='rsem_ref' , desc='RSEM Reference Name', readonly=True, visible=True, set_in_upload=True, no_value='rsem_ref' )
54 file_ext = 'rsem_ref'
55 allow_datatype_change = False
56 composite_type = 'auto_primary_file'
57 def __init__( self, **kwd ):
58 Html.__init__(self, **kwd)
59 """
60 Expecting files:
61 extra_files_path/<reference_name>.grp
62 extra_files_path/<reference_name>.ti
63 extra_files_path/<reference_name>.seq
64 extra_files_path/<reference_name>.transcripts.fa
65 Optionally includes files:
66 extra_files_path/<reference_name>.chrlist
67 extra_files_path/<reference_name>.idx.fa
68 extra_files_path/<reference_name>.1.ebwt
69 extra_files_path/<reference_name>.2.ebwt
70 extra_files_path/<reference_name>.3.ebwt
71 extra_files_path/<reference_name>.4.ebwt
72 extra_files_path/<reference_name>.rev.1.ebwt
73 extra_files_path/<reference_name>.rev.2.ebwt
74 """
75 self.add_composite_file( '%s.grp', description = 'Group File', substitute_name_with_metadata = 'reference_name', is_binary = False )
76 self.add_composite_file( '%s.ti', description = '', substitute_name_with_metadata = 'reference_name', is_binary = False )
77 self.add_composite_file( '%s.seq', description = '', substitute_name_with_metadata = 'reference_name', is_binary = False )
78 self.add_composite_file( '%s.transcripts.fa', description = '', substitute_name_with_metadata = 'reference_name', is_binary = False )
79 self.add_composite_file( '%s.chrlist', description = '', substitute_name_with_metadata = 'reference_name', is_binary = False, optional=True )
80 self.add_composite_file( '%s.idx.fa', description = '', substitute_name_with_metadata = 'reference_name', is_binary = False, optional=True )
81 self.add_composite_file( '%s.1.ebwt', description = '', substitute_name_with_metadata = 'reference_name', is_binary = True, optional=True )
82 self.add_composite_file( '%s.2.ebwt', description = '', substitute_name_with_metadata = 'reference_name', is_binary = True, optional=True )
83 self.add_composite_file( '%s.3.ebwt', description = '', substitute_name_with_metadata = 'reference_name', is_binary = True, optional=True )
84 self.add_composite_file( '%s.4.ebwt', description = '', substitute_name_with_metadata = 'reference_name', is_binary = True, optional=True )
85 self.add_composite_file( '%s.rev.1.ebwt', description = '', substitute_name_with_metadata = 'reference_name', is_binary = True, optional=True )
86 self.add_composite_file( '%s.rev.2.ebwt', description = '', substitute_name_with_metadata = 'reference_name', is_binary = True, optional=True )
87
88 def generate_primary_file( self, dataset = None ):
89 """
90 This is called only at upload to write the file
91 cannot rename the datasets here - they come with the default unfortunately
92 """
93
94 def regenerate_primary_file(self,dataset):
95 """
96 cannot do this until we are setting metadata
97 """
98 link_to_exts = ['.grp','.ti','.seq','.fa','.chrlist','.log']
99 ref_name = dataset.metadata.reference_name
100 efp = dataset.extra_files_path
101 flist = os.listdir(efp)
102 rval = ['<html><head><title>%s</title></head><body><p/>RSEM Reference %s files:<p/><ul>' % (dataset.name,ref_name)]
103 rvalb = []
104 for i,fname in enumerate(flist):
105 sfname = os.path.split(fname)[-1]
106 f,e = os.path.splitext(fname)
107 if e in link_to_exts:
108 rval.append( '<li><a href="%s">%s</a></li>' % ( sfname, sfname) )
109 else:
110 rvalb.append( '<li>%s</li>' % (sfname) )
111 if len(rvalb) > 0:
112 rval += rvalb
113 rval.append( '</ul></body></html>' )
114 fh = file(dataset.file_name,'w')
115 fh.write("\n".join( rval ))
116 fh.write('\n')
117 fh.close()
118
119 def set_meta( self, dataset, **kwd ):
120 Html.set_meta( self, dataset, **kwd )
121 efp = dataset.extra_files_path
122 flist = os.listdir(efp)
123 for i,fname in enumerate(flist):
124 if fname.endswith('.grp'):
125 dataset.metadata.reference_name = fname[:-4]
126 break
127 self.regenerate_primary_file(dataset)
128