Add filtering for rnafusion

.groovylintrc.json

@@ -9,6 +9,9 @@
     },
     "DuplicateStringLiteral": {
       "enabled": false
+    },
+    "UnnecessaryGetter": {
+      "enabled": false
     }
   }
 }

CHANGELOG.md

@@ -1,3 +1,8 @@
+## 0.2.4
+
+1. Fixed samplesheets being empty in 0.2.3
+2. Added a filter for rnafusion samplesheets. Samples with less than 1 million reads will now be put in a `nfcore_rnafusion_samplesheet_failed.yaml` file
+
 ## 0.2.3
 
 Added `tissue` samples to the vivar samplesheet

build.gradle

@@ -6,7 +6,7 @@ dependencies {
     implementation 'com.squareup.okhttp3:okhttp:5.3.2'
 }
 
-version = '0.2.3'
+version = '0.2.4'
 
 nextflowPlugin {
     nextflowVersion = '25.10.0'

src/main/groovy/nfcmgg/plugin/samplesheets/PipelineObserver.groovy

@@ -95,12 +95,10 @@ class PipelineObserver implements TraceObserverV2 {
      * @return the sample name for the given base path
      */
     protected String safeGetSample(String inputBasePath) {
-        String basePath
+        String basePath = inputBasePath
         // Make sure SNP tracking data will be added to the correct sample
         if (basePath.startsWith('snp_')) {
             basePath = inputBasePath.replaceFirst('snp_', '')
-        } else {
-            basePath = inputBasePath
         }
         List<String> possibleSamples = samples
             .findAll { sample -> basePath.startsWith(sample) }.toList() as List<String>

src/main/groovy/nfcmgg/plugin/samplesheets/PreprocessingObserver.groovy

@@ -38,6 +38,23 @@ class PreprocessingObserver extends PipelineObserver {
     void onFilePublish(FilePublishEvent event) {
         String targetName = event.target.name
         String targetPath = event.target.toUriString()
+
+        // Get metrics after demultiplexing
+        if (targetPath.endsWith('_SAV_data') && event.source.isDirectory()) {
+            Path bysamples = event.source.resolve('multiqc_bclconvert_bysample.txt')
+            if (bysamples.exists()) {
+                List<Map> sampleMetrics = bysamples.splitCsv(header:true, sep:'\t')
+                sampleMetrics.each { metric ->
+                    String sample = metric['Sample']
+                    String yield = metric['yield_']
+                    if (entries.containsKey(sample)) {
+                        entries[sample].add('yield', yield)
+                    }
+                }
+            }
+        }
+
+        // Get specific files
         switch (targetName) {
             case ~/^snp_.*\.cram$/:
                 entries[safeGetSample(targetName)].add('snp_cram', targetPath)
@@ -103,24 +120,44 @@ class PreprocessingObserver extends PipelineObserver {
         //
         // nf-core/rnafusion samplesheet
         //
+
+        Map<String, OutputEntry> rnafusionEntries = entries
+            .findAll { entry ->
+                // Only create samplesheet for RNAseqMDG runs of RNA samples that have FASTQ output
+                entry.value.get('sample_type')?.toLowerCase() == 'rna' &&
+                entry.value.get('tag')?.toLowerCase() == 'rnaseqmdg' &&
+                entry.value.get('fastq_1')
+            }
+        List rnafusionKeys = [
+            ['id', 'sample'],
+            'fastq_1',
+            'fastq_2',
+            'strandedness',
+            ['yield', 'reads']
+        ]
+
+        // Passed data
         creator.dump(
-            entries
+            rnafusionEntries
                 .findAll { entry ->
-                    // Only create samplesheet for RNAseqMDG runs of RNA samples that have FASTQ output
-                    entry.value.get('sample_type')?.toLowerCase() == 'rna' &&
-                    entry.value.get('tag')?.toLowerCase() == 'rnaseqmdg' &&
-                    entry.value.get('fastq_1')
+                    entry.value.get('yield').toInteger() >= 1000000
                 }
                 .values()
-                *.subKeys([
-                    ['id', 'sample'],
-                    'fastq_1',
-                    'fastq_2',
-                    'strandedness'
-                ]),
+                *.subKeys(rnafusionKeys),
             location.resolve('nfcore_rnafusion_samplesheet.yaml')
         )
 
+        // Failed data
+        creator.dump(
+            rnafusionEntries
+                .findAll { entry ->
+                    entry.value.get('yield').toInteger() < 1000000
+                }
+                .values()
+                *.subKeys(rnafusionKeys),
+            location.resolve('nfcore_rnafusion_samplesheet_failed.yaml')
+        )
+
         //
         // nf-cmgg/vivar samplesheet
         //
@@ -209,6 +246,7 @@ class PreprocessingObserver extends PipelineObserver {
             'library': sampleData.get('library', null),
             'sex': sampleData.get('sex', 'U'),
             'exomecnv_batch': (sampleData.get('library', null) as String) + '_' + sampleData.get('sex', 'U'),
+            'yield': -1
         ]
     }