Changeset 24 for trunk/grails-app/services/nl

Timestamp:

Mar 23, 2011, 1:24:24 PM (12 years ago)

Author:

robert@…

Message:

Improved export of fasta files and added properties to assaysamples

Location:

trunk/grails-app/services/nl/tno/metagenomics

Files:

• FastaService.groovy (modified) (8 diffs)
• SampleExcelService.groovy (modified) (11 diffs)
• files/ExcelService.groovy (modified) (1 diff)
• integration/SynchronizationService.groovy (modified) (4 diffs)

trunk/grails-app/services/nl/tno/metagenomics/FastaService.groovy

@@ -405 +405 @@
          * @return
          */
-        public def export( List assaySamples, OutputStream outStream, String name ) {
+        public def export( List assaySamples, OutputStream outStream, String name = null ) {
                 if( !assaySamples || assaySamples.size() == 0 )
                         return false;
+                
+                // Retrieve the filename from configuration, if none is given
+                if( !name ) 
+                        name = ConfigurationHolder.config.metagenomics.exportFilename
 
                 // Determine the directory the uploaded files are stored in
@@ -421 +425 @@
                 // Determine new tag length. Since we can use 4 characters per
                 // tag position, we only have to use 4log( #samples)
-                int tagLength = Math.ceil( Math.log( assaySamples.size() ) / Math.log( 4 ) )
+                // The minimum number of characters used is 10, to ensure the correct working of the other
+                // programs.
+                int tagLength = Math.max( 10, Math.ceil( Math.log( assaySamples.size() ) / Math.log( 4 ) ) );
                 int tagNumber = 0;
 
@@ -430 +436 @@
 
                                 // Save the tag for exporting
-                                tags << [assaySampleId: assaySample.id, sampleName: assaySample.sample.name, assayName: assaySample.assay.name, studyName: assaySample.assay.study.name, tag: tag]
+                                tags << [       assaySampleId: assaySample.id, sampleName: assaySample.sample.name,
+                                                        assayName: assaySample.assay.name, studyName: assaySample.assay.study.name,
+                                                        forwardPrimer: assaySample.fwPrimerSeq, reversePrimer: assaySample.revPrimerSeq, 
+                                                        tag: tag
+                                ];
                         }
                 }
@@ -441 +451 @@
                 // file and part of the qual files mixed. We have to write the full sequence file first.
                 try {
-                        zipFile.putNextEntry( new ZipEntry( name + ".fasta" ) );
+                        zipFile.putNextEntry( new ZipEntry( name + ".fna" ) );
                         
                         assaySamples.each { assaySample ->
@@ -484 +494 @@
                 
                 // Export a tab delimited file with tags
-                zipFile.putNextEntry( new ZipEntry( name + "_sample-tag.tab" ) );
-                exportSampleTagFile( tags, zipWriter );
+                zipFile.putNextEntry( new ZipEntry( name + ".tab" ) );
+                exportTabDelimitedSampleTagFile( tags, zipWriter );
                 zipWriter.flush();
                 zipFile.closeEntry();
 
+                // Export a mothur file with tags
+                zipFile.putNextEntry( new ZipEntry( name + ".oligos" ) );
+                exportMothurSampleTagFile( tags, zipWriter );
+                zipWriter.flush();
+                zipFile.closeEntry();
+                
                 // Export an excel file with information about the samples
                 zipFile.putNextEntry( new ZipEntry( name + ".xls" ) );
@@ -498 +514 @@
         
         /**
-         * Creates a tab delimited file with two columns and column headers "Sequence" and "Samplename"
+         * Creates an oligos file for Mothur that represents the connection between samples
+         * and the artificial tags. 
+         * 
+         * @see http://www.mothur.org/wiki/Trim.seqs#allfiles
          * @param tags          Map with newly created tags
          * @param zipWriter     Writer to write the data to
          */
-        protected void exportSampleTagFile( List tags, Writer zipWriter ) {
-                zipWriter.write( "Sequence" + "\t" + "Samplename" + "\n" );
+        protected void exportMothurSampleTagFile( List tags, Writer zipWriter ) {
+                // Add the forward and reverse primers, as found in the assaysamples
+                // The primers are already cut off, so they are not relevant anymore
+                // For that reason, a '#' is prepended to each line.
+                def fwPrimers = tags.collect { it.forwardPrimer }.findAll { it }.unique();
+                def revPrimers = tags.collect { it.reversePrimer }.findAll { it }.unique();
+                
+                fwPrimers.each { zipWriter.write( "#forward\t" + it + "\n" )    }
+                revPrimers.each { zipWriter.write( "#reverse\t" + it + "\n" ) }
                 
                 // Check whether the sample names are unique. If they aren't, the assay and study names
@@ -519 +545 @@
                 
                 tags.each {
-                        zipWriter.write( it.tag + "\t" + it.uniqueName + "\n" );
-                }
-        }
-
+                        zipWriter.write( "barcode\t" + it.tag + "\t" + it.uniqueName + "\n" );
+                }
+        }
+
+        /**
+        * Creates a tab delimited file with two columns and column headers "Sequence" and "Samplename"
+        * @param tags           Map with newly created tags
+        * @param zipWriter      Writer to write the data to
+        */
+   protected void exportTabDelimitedSampleTagFile( List tags, Writer zipWriter ) {
+           zipWriter.write( "Sequence" + "\t" + "Samplename" + "\n" );
+           
+           // Check whether the sample names are unique. If they aren't, the assay and study names
+           // are appended to the sample name
+           def sampleNames = tags*.sampleNames;
+           if( sampleNames.unique().size() < sampleNames.size() ) {
+                   tags.each {
+                           it.uniqueName = it.sampleName + " (" + it.assayName + " / " + it.studyName + ")";
+                   }
+           } else {
+                   tags.each {
+                           it.uniqueName = it.sampleName;
+                   }
+           }
+           
+           tags.each {
+                   zipWriter.write( it.tag + "\t" + it.uniqueName + "\n" );
+           }
+   }
+
+        
         /**
          * Creates a unique tag for the given number
@@ -541 +594 @@
                         int currentChar = tagNumber % numChars
 
-                        tag = chars[ currentChar ] + tag;
+                        // Append the new character to the end of the tag, to ensure that the first part of the tag is
+                        // the most volatile. This way it is easy to find the end of the tag and the beginning of the real
+                        // sequence on first sight.
+                        tag = tag + chars[ currentChar ];
 
                         tagNumber = Math.floor( tagNumber / numChars );
                 }
 
-                return tag
+                return tag;
         }
 

trunk/grails-app/services/nl/tno/metagenomics/SampleExcelService.groovy

@@ -11 +11 @@
 
         // Fields to be edited using excel file and manually
+        def variableFields = [
+                'fwOligo':              'Forward oligo number',
+                'fwMidName':    'Forward mid name',
+                'fwTotalSeq':   'Total forward sequence',
+                'fwMidSeq':             'Forward mid sequence',
+                'fwPrimerSeq':  'Forward primer sequence',
+                'revOligo':     'Reverse oligo number',
+                'revMidName':   'Reverse mid name',
+                'revTotalSeq':  'Total reverse sequence',
+                'revMidSeq':    'Reverse mid sequence',
+                'revPrimerSeq': 'Reverse primer sequence',
+
+        ]
+        
         def sampleNameName = "Sample name"
         def runName = "Run"
-        def tagNameName = "Tag name"
-        def tagSequenceName = "Tag sequence"
-        def oligoNumberName = "Oligo number"
-        def possibleFields = [sampleNameName, runName, tagNameName, tagSequenceName, oligoNumberName]
+        def possibleFields = [sampleNameName, runName] + variableFields.keySet().toList();
+        def possibleFieldNames = [sampleNameName, runName ] + variableFields.values();
         
     /**
@@ -35 +47 @@
                 def wb = excelService.create();
 
-                def fields = possibleFields
+                def fields = possibleFieldNames
                 if( !includeRun )
                         fields = fields - runName
@@ -49 +61 @@
                                 rowData << assaySample.run?.name
                         
-                        rowData << assaySample.tagName
-                        rowData << assaySample.tagSequence
-                        rowData << assaySample.oligoNumber
+                        variableFields.each { k, v ->
+                                rowData << assaySample[ k ];
+                        }
+                        
                         data << rowData;
                 }
@@ -57 +70 @@
 
                 // Auto resize columns
-                excelService.autoSizeColumns( wb, sheetIndex, 0..2)
+                excelService.autoSizeColumns( wb, sheetIndex, 0..fields.size())
 
                 return wb;
@@ -82 +95 @@
                 // Try to guess best matches between the excel file and the column names
                 def bestMatches = [:]
-                def fields = possibleFields
+                def fields = possibleFieldNames
                 if( !includeRun )
                         fields = fields - runName
-                
                         
                 // Do matching using fuzzy search. The 0.8 treshold makes sure that no match if chosen if
                 // there is actually no match at all.
-                def matches = fuzzySearchService.mostSimilarUnique( headers, possibleFields, 0.8 );
+                def matches = fuzzySearchService.mostSimilarUnique( headers, fields, 0.8 );
                 
                 headers.eachWithIndex { header, idx ->
@@ -121 +133 @@
                 def columns = [:]
                 def dataMatches = false;
-                possibleFields.each { columnName ->
+                possibleFieldNames.each { columnName ->
                         def foundColumn = matchColumns.find { it.value == columnName };
                         
@@ -130 +142 @@
                 }
 
-                println columns
-                
                 // A column to match the sample name must be present
                 if( columns[ sampleNameName ] == -1 ) {
@@ -188 +198 @@
                         columns.each {
                                 if( it.value > -1 ) {
-                                        switch( it.key ) {
-                                                case tagNameName:               assaySample.tagName = rowData[ it.value ]; break
-                                                case tagSequenceName:   assaySample.tagSequence = rowData[ it.value ]; break
-                                                case oligoNumberName:   assaySample.oligoNumber = rowData[ it.value ]; break
-                                                case runName:                   assaySample.run = Run.findByName( rowData[ it.value ] ); break
+                                        if( it.key == runName ) {
+                                                assaySample.run = Run.findByName( rowData[ it.value ] );
+                                        } else {
+                                                def field = variableFields.find { variableField -> variableField.value == it.key }; 
+                                                if( field ) {
+                                                        assaySample[ field.key ] = rowData[ it.value ];
+                                                }
                                         }
                                 }
@@ -245 +257 @@
                 def subjectFields = []
                 def eventFields = []
-                def moduleFields = [ "Sample name", "Assay name", "Study name", "Run name", "# sequences", "Artificial tag sequence", "Original tag sequence", "Tag name", "Oligo number" ]
+                def moduleFields = [ "Sample name", "Assay name", "Study name", "Run name", "# sequences", "Artificial tag sequence" ] + variableFields.values();
                 gscfData.each { sample ->
                         sample.each { key, value ->
@@ -281 +293 @@
                                 assaySample.numSequences(),
                                 currentTag?.tag,
-                                assaySample.tagName,
-                                assaySample.tagSequence,
-                                assaySample.oligoNumber
                         ]
+                        
+                        // Add the variable fields for all assaysamples
+                        variableFields.each { k, v ->
+                                row << assaySample[ k ];
+                        }
                         
                         // Afterwards add the gscfData including subject and event data
@@ -322 +336 @@
                 excelService.writeData( wb, data, sheetIndex, 0 );
 
-                // Auto resize columns
-                excelService.autoSizeColumns( wb, sheetIndex, 0..fields[ "names" ][ "all" ].size()-1)
+                // Auto resize columns (# columns = # samples + 1)
+                excelService.autoSizeColumns( wb, sheetIndex, 0..assaySamples?.size())
 
                 // Write the data to the output stream

trunk/grails-app/services/nl/tno/metagenomics/files/ExcelService.groovy

@@ -185 +185 @@
          * @return
          */
-        public Workbook autoSizeColumns( Workbook book, int sheetIndex = 0,def columns = 0 ) {
+        public Workbook autoSizeColumns( Workbook book, int sheetIndex = 0, def columns = 0 ) {
                 if( book == null )
                         throw new Exception( "No workbook given." );

trunk/grails-app/services/nl/tno/metagenomics/integration/SynchronizationService.groovy

@@ -634 +634 @@
                                         // Update the sample object if necessary
                                         sampleFound.name = gscfSample.name
-                                        sampleFound.subject = gscfSample.subject.toString()
-                                        sampleFound.event = gscfSample.event.toString() + ( gscfSample.startTime ? " (" + gscfSample.startTime + ")" : "" )
+                                        setSubjectAndEventFromGSCF( sampleFound, gscfSample );
                                         sampleFound.save();
                                 } else {
@@ -648 +647 @@
                                                 // Update the sample object if necessary
                                                 sampleFound.name = gscfSample.name
-                                                sampleFound.subject = gscfSample.subject.toString()
-                                                sampleFound.event = gscfSample.event.toString() + ( gscfSample.startTime ? " (" + gscfSample.startTime + ")" : "" )
+                                                setSubjectAndEventFromGSCF( sampleFound, gscfSample );
                                                 sampleFound.save();
                                         } else {
@@ -656 +654 @@
                                                 // If it doesn't exist, create a new object
                                                 sampleFound = new Sample( sampleToken: gscfSample.sampleToken, name: gscfSample.name, study: assay.study );
-                                                sampleFound.subject = gscfSample.subject.toString()
-                                                sampleFound.event = gscfSample.event.toString() + ( gscfSample.startTime ? " (" + gscfSample.startTime + ")" : "" )
+                                                setSubjectAndEventFromGSCF( sampleFound, gscfSample );
                                                 assay.study.addToSamples( sampleFound );
                                                 sampleFound.save();
@@ -673 +670 @@
                         }
                 }
+        }
+        
+        /**
+         * Copies the subject and event properties from the gscf sample to the local sample
+         * @param sample                Sample object to update
+         * @param gscfSample    Map with properties about the gscf sample ('subject', 'event' and 'startTime' are used)
+         */
+        private void setSubjectAndEventFromGSCF( sample, gscfSample ) {
+                sample.subject = gscfSample.subject && gscfSample.subject != "null" ? gscfSample.subject.toString() : ""
+
+                sample.event = gscfSample.event && gscfSample.event != "null" ? gscfSample.event.toString() : ""
+                
+                if( gscfSample.startTime && gscfSample.startTime != "null" ) 
+                        sample.event += " (" + gscfSample.startTime + ")";
         }