changeset 5:503097df1670 draft

Uploaded

--- a/data_manager/rnastar_index_builder.py	Thu Jan 08 17:53:14 2015 -0500
+++ b/data_manager/rnastar_index_builder.py	Fri Jan 09 22:38:21 2015 -0500
@@ -2,17 +2,25 @@
 #Dan Blankenberg
 # adapted from Dan's BWA one for rna star
 # ross lazarus sept 2014
-#
+# fixed some stupid bugs January 2015
+
+
 import sys
 import os
 import tempfile
 import optparse
 import subprocess
 
-from json import loads, dumps
+from galaxy.util.json import from_json_string, to_json_string
+
+
+CHUNK_SIZE = 2**20
+ONE_GB = 2**30
+
+# from json import loads, dumps
 
 DEFAULT_DATA_TABLE_NAME = "rnastar_index"
-CHUNK_SIZE = 2**20
+
 
 def get_id_name( params, dbkey, fasta_description=None):
     #TODO: ensure sequence_id is unique and does not already appear in location file
@@ -50,10 +58,10 @@
     sjdbOverhang,sjdbGTFfile, sjdbFileChrStartEnd,sjdbGTFtagExonParentTranscript,sjdbGTFfeatureExon,sjdbGTFchrPrefix,n_threads):
     #TODO: allow multiple FASTA input files
     fasta_base_name = os.path.basename( fasta_filename )
-    #sym_linked_fasta_filename = os.path.join( tmp_dir, fasta_base_name )
-    #os.symlink( fasta_filename, sym_linked_fasta_filename )
-    #print >> sys.stdout,'made',sym_linked_fasta_filename
-    cl = ['STAR','--runMode', 'genomeGenerate', '--genomeFastaFiles', fasta_filename, '--genomeDir', target_directory, '--runThreadN',n_threads ]
+    sym_linked_fasta_filename = os.path.join( target_directory, fasta_base_name )
+    os.symlink( fasta_filename, sym_linked_fasta_filename )
+    print >> sys.stdout,'made',sym_linked_fasta_filename
+    cl = ['STAR','--runMode', 'genomeGenerate', '--genomeFastaFiles', sym_linked_fasta_filename, '--genomeDir', target_directory, '--runThreadN',n_threads ]
     
     if sjdbGTFfile:
          cl += [ '--sjdbGTFfeatureExon', sjdbGTFfeatureExon,'--sjdbGTFtagExonParentTranscript',sjdbGTFtagExonParentTranscript]
@@ -62,14 +70,31 @@
          cl += ['--sjdbOverhang', sjdbOverhang, '--sjdbGTFfile', sjdbGTFfile]
     elif sjdbFileChrStartEnd:
         cl += ['--sjdbFileChrStartEnd', sjdbFileChrStartEnd, '--sjdbOverhang', sjdbOverhang]
-    return_code = _run_command(command=cl,target_directory=target_directory)
+
+    tmp_stderr = tempfile.NamedTemporaryFile( prefix = "tmp-data-manager-rnastar-index-builder-stderr" )
+    proc = subprocess.Popen( args=cl, shell=False, cwd=target_directory, stderr=tmp_stderr.fileno() )
+    return_code = proc.wait()
+    if return_code:
+        tmp_stderr.flush()
+        tmp_stderr.seek(0)
+        print >> sys.stderr, "Error building index:"
+        while True:
+            chunk = tmp_stderr.read( CHUNK_SIZE )
+            if not chunk:
+                break
+            sys.stderr.write( chunk )
+        sys.exit( return_code )
+    tmp_stderr.close()
     data_table_entry = dict( value=sequence_id, dbkey=dbkey, name=sequence_name, path=fasta_base_name )
-    data_manager_dict.setdefault('data_tables',{})
-    data_manager_dict['data_tables'].setdefault(data_table_name,[])
+    _add_data_table_entry( data_manager_dict, data_table_name, data_table_entry )
+
+
+def _add_data_table_entry( data_manager_dict, data_table_name, data_table_entry ):
+    data_manager_dict['data_tables'] = data_manager_dict.get( 'data_tables', {} )
+    data_manager_dict['data_tables'][ data_table_name ] = data_manager_dict['data_tables'].get( data_table_name, [] )
     data_manager_dict['data_tables'][ data_table_name ].append( data_table_entry )
     return data_manager_dict
-
-
+    
 def main():
     #Parse Command Line
     parser = optparse.OptionParser()
@@ -85,34 +110,28 @@
     parser.add_option( '--sjdbOverhang', type="string", default='100' )
     parser.add_option( '--runThreadN', type="string", default='4' )
     (options, args) = parser.parse_args()
-    filename = args[0]    
-    # this is passed as an positional parameter in the original wrapper with the extra_files_path from the params
-    params = loads( open( filename ).read() )
+    filename = args[0]
+    params = from_json_string( open( filename ).read() )
     target_directory = params[ 'output_data' ][0]['extra_files_path'].encode('ascii','replace')
+    data_manager_dict = {}
+    dbkey = options.fasta_dbkey
+    if dbkey in [ None, '', '?' ]:
+        raise Exception( '"%s" is not a valid dbkey. You must specify a valid dbkey.' % ( dbkey ) )
+    
+    sequence_id, sequence_name = get_id_name( params, dbkey=dbkey, fasta_description=options.fasta_description )
+    
     try:
          os.mkdir( target_directory )
     except OSError:
         pass
-
-    data_manager_dict = {}
-    
-    dbkey = options.fasta_dbkey
-    
-    if dbkey in [ None, '', '?' ]:
-        raise Exception( 'dbkey "%s" does not specify any locally known genome to be indexed.' % ( dbkey ) )
-    
-    sequence_id, sequence_name = get_id_name( params, dbkey=dbkey, fasta_description=options.fasta_description )
-    
     #build the index
-    data_manager_dict = build_rnastar_index( data_manager_dict = data_manager_dict, fasta_filename = options.fasta_filename, target_directory = target_directory,
+    build_rnastar_index( data_manager_dict = data_manager_dict, fasta_filename = options.fasta_filename, target_directory = target_directory,
       dbkey = dbkey, sequence_id = sequence_id, sequence_name = sequence_name, data_table_name=options.data_table_name,
       sjdbOverhang=options.sjdbOverhang,sjdbGTFfile=options.sjdbGTFfile,
       sjdbFileChrStartEnd=options.sjdbFileChrStartEnd,sjdbGTFtagExonParentTranscript=options.sjdbGTFtagExonParentTranscript,
       sjdbGTFfeatureExon=options.sjdbGTFfeatureExon,sjdbGTFchrPrefix=options.sjdbGTFchrPrefix,
       n_threads=options.runThreadN )
-
-    
-    #save info to json file
-    open( filename, 'wb' ).write( dumps( data_manager_dict ) )
+    open( filename, 'wb' ).write( to_json_string( data_manager_dict ) )
         
 if __name__ == "__main__": main()
+

--- a/data_manager/rnastar_index_builder.xml	Thu Jan 08 17:53:14 2015 -0500
+++ b/data_manager/rnastar_index_builder.xml	Fri Jan 09 22:38:21 2015 -0500
@@ -1,4 +1,4 @@
-<tool id="rnastar_index_builder_data_manager" name="rnastar index" tool_type="manage_data" version="0.0.1">
+<tool id="rnastar_index_builder_data_manager" name="rnastar index" tool_type="manage_data" version="0.0.2">
     <description>builder</description>
     <requirements>
         <requirement type="package" version="2.4.0d">rnastar</requirement>
@@ -23,18 +23,12 @@
 #end if
 --data_table_name "rnastar_index" 
 </command>
-
-    <stdio>
-        <regex match=".*" source="both" level="warning" description="stdout/err chatter:"/>
-    </stdio>
     <inputs>
         <param name="all_fasta_source" type="select" label="Source FASTA Sequence">
             <options from_data_table="all_fasta"/>
         </param>
         <param type="text" name="sequence_name" value="" label="Informative name for sequence index" />
         <param type="text" name="sequence_id" value="" label="ID for sequence index" />
-
-
         <conditional name="genemodel">
             <param name="modelformat" type="select" 
                label="Choose the format of gene model data from your history - bed or gff3"
@@ -68,14 +62,14 @@
                 </param>
                    
                 <param type="integer" name="sjdbOverhang" value="100" label="Splice junction overhang. If=0, splice junction database NOT used"
-                help="int>=0: length of the donor/acceptor sequence on each side, (mate_length - 1)" />
+                help="integer length of the donor/acceptor sequence on each side, (mate_length - 1)" />
                 
             </when>
             <when value='bed'>
                 <param type="data" format="bed" name="sjdbFileChrStartEnd" value="" label="Introns as a tabular bed (chr,start,end,strand) file matching the input genome"
                    help="Required if you want to index splice junctions during index generation." />
                 <param type="integer" name="sjdbOverhang" value="100" label="Splice junction overhang. If=0, splice junction database NOT used"
-                help="int>=0: length of the donor/acceptor sequence on each side, (mate_length - 1)" />
+                help="integer length of the donor/acceptor sequence on each side, (mate_length - 1)" />
            </when>
             <when value='None'>
             </when>