Merge pull request #166 from biopet/fix-BIOPET-718

Removing sample/library variantcalling

biopet-core/src/main/scala/nl/lumc/sasc/biopet/core/summary/WriteSummary.scala

@@ -177,6 +177,11 @@ class WriteSummary(val parent: SummaryQScript) extends InProcessFunction with Co
         val libId = tag.libId.flatMap(name =>
           sampleId.flatMap(sampleId =>
             Await.result(db.getLibraryId(qscript.summaryRunId, sampleId, name), Duration.Inf)))
+        if (tag.sampleId.isDefined)
+          require(sampleId.isDefined, s"Sample '${tag.sampleId.get}' is not found in database yet")
+        if (tag.libId.isDefined)
+          require(libId.isDefined,
+                  s"Library '${tag.libId.get}' for '${tag.sampleId}' is not found in database yet")
         for ((key, file) <- qscript.summaryFiles.par)
           Await.result(WriteSummary.createFile(db,
                                                qscript.summaryRunId,
@@ -353,7 +358,10 @@ object WriteSummary {
 
   /** Retrive checksum from file */
   def parseChecksum(checksumFile: File): String = {
-    Source.fromFile(checksumFile).getLines().toList.head.split(" ")(0)
+    val reader = Source.fromFile(checksumFile)
+    val lines = reader.getLines().toList
+    reader.close()
+    lines.head.split(" ")(0)
   }
 
   def createFile(db: SummaryDbWrite,

biopet-utils/src/main/scala/nl/lumc/sasc/biopet/utils/summary/db/Schema.scala

@@ -110,8 +110,6 @@ object Schema {
 
     def * =
       (runId, pipelineId, moduleId, sampleId, libraryId, content) <> (Stat.tupled, Stat.unapply)
-
-    def idx = index("idx_stats", (runId, pipelineId, moduleId, sampleId, libraryId), unique = true)
   }
   val stats = TableQuery[Stats]
 
@@ -131,9 +129,6 @@ object Schema {
 
     def * =
       (runId, pipelineId, moduleId, sampleId, libraryId, content) <> (Setting.tupled, Setting.unapply)
-
-    def idx =
-      index("idx_settings", (runId, pipelineId, moduleId, sampleId, libraryId), unique = true)
   }
   val settings = TableQuery[Settings]
 
@@ -163,7 +158,7 @@ object Schema {
       (runId, pipelineId, moduleId, sampleId, libraryId, key, path, md5, link, size) <> (File.tupled, File.unapply)
 
     def idx =
-      index("idx_files", (runId, pipelineId, moduleId, sampleId, libraryId, key), unique = true)
+      index("idx_files", (runId, pipelineId, moduleId, sampleId, libraryId, key))
   }
   val files = TableQuery[Files]
 

shiva/src/main/scala/nl/lumc/sasc/biopet/pipelines/shiva/Shiva.scala

@@ -101,17 +101,13 @@ class Shiva(val parent: Configurable)
     /** Sample specific settings */
     override def summarySettings: Map[String, Any] =
       super.summarySettings ++
-        Map(
-          "single_sample_variantcalling" -> variantcalling.isDefined,
-          "use_indel_realigner" -> useIndelRealigner
-        )
+        Map("use_indel_realigner" -> useIndelRealigner)
 
     /** Class to generate jobs for a library */
     class Library(libId: String) extends super.Library(libId) {
 
       override def summaryFiles: Map[String, File] =
         super.summaryFiles ++
-          variantcalling.map("final" -> _.finalFile) ++
           bqsrFile.map("baserecal" -> _) ++
           bqsrAfterFile.map("baserecal_after" -> _)
 
@@ -140,20 +136,10 @@ class Shiva(val parent: Configurable)
 
       /** Library specific settings */
       override def summarySettings: Map[String, Any] = super.summarySettings ++ Map(
-        "library_variantcalling" -> variantcalling.isDefined,
         "use_base_recalibration" -> useBaseRecalibration,
         "useAnalyze_covariates" -> useAnalyzeCovariates
       )
 
-      lazy val variantcalling: Option[ShivaVariantcalling with QScript] =
-        if (config("library_variantcalling", default = false).asBoolean &&
-            (bamFile.isDefined || preProcessBam.isDefined)) {
-          Some(
-            makeVariantcalling(multisample = false,
-                               sample = Some(sampleId),
-                               library = Some(libId)))
-        } else None
-
       /** This will add jobs for this library */
       override def addJobs(): Unit = {
         super.addJobs()
@@ -164,15 +150,6 @@ class Shiva(val parent: Configurable)
                               useIndelRealigner || libraries.size > 1,
                               usePrintReads)
         }
-
-        variantcalling.foreach(vc => {
-          vc.sampleId = Some(sampleId)
-          vc.libId = Some(libId)
-          vc.outputDir = new File(libDir, "variantcalling")
-          if (preProcessBam.isDefined) vc.inputBams = Map(sampleId -> preProcessBam.get)
-          else vc.inputBams = Map(sampleId -> bamFile.get)
-          add(vc)
-        })
       }
 
       /** Adds base recalibration jobs */
@@ -220,11 +197,6 @@ class Shiva(val parent: Configurable)
       }
     }
 
-    lazy val variantcalling: Option[ShivaVariantcalling with QScript] =
-      if (config("single_sample_variantcalling", default = false).asBoolean) {
-        Some(makeVariantcalling(multisample = false, sample = Some(sampleId)))
-      } else None
-
     override def keepMergedFiles: Boolean =
       config("keep_merged_files", default = !useIndelRealigner)
 
@@ -235,9 +207,6 @@ class Shiva(val parent: Configurable)
         bamFile.map(swapExt(sampleDir, _, ".bam", ".realign.bam"))
       } else bamFile
 
-    override def summaryFiles: Map[String, File] =
-      super.summaryFiles ++ variantcalling.map("final" -> _.finalFile)
-
     /** This will add sample jobs */
     override def addJobs(): Unit = {
       super.addJobs()
@@ -245,15 +214,6 @@ class Shiva(val parent: Configurable)
       if (useIndelRealigner) {
         addIndelRealign(bamFile.get, sampleDir, isIntermediate = false)
       }
-
-      preProcessBam.foreach { bam =>
-        variantcalling.foreach(vc => {
-          vc.sampleId = Some(sampleId)
-          vc.outputDir = new File(sampleDir, "variantcalling")
-          vc.inputBams = Map(sampleId -> bam)
-          add(vc)
-        })
-      }
     }
   } // End of sample
 

shiva/src/test/scala/nl/lumc/sasc/biopet/pipelines/shiva/ShivaTest.scala

@@ -60,8 +60,6 @@ trait ShivaTestTrait extends TestNGSuite with Matchers {
   def realignProvider = Array(false, true)
   def baseRecalibrationProvider = Array(false, true)
   def multisampleCalling: Boolean = true
-  def sampleCalling = false
-  def libraryCalling = false
   def dbsnp = true
   def svCalling = false
   def cnvCalling = false
@@ -86,8 +84,6 @@ trait ShivaTestTrait extends TestNGSuite with Matchers {
       ConfigUtils.mergeMaps(
         Map(
           "multisample_variantcalling" -> multisampleCalling,
-          "single_sample_variantcalling" -> sampleCalling,
-          "library_variantcalling" -> libraryCalling,
           "use_indel_realigner" -> realign,
           "use_base_recalibration" -> baseRecalibration,
           "sv_calling" -> svCalling,
@@ -133,26 +129,24 @@ trait ShivaTestTrait extends TestNGSuite with Matchers {
 
       pipeline.samples foreach {
         case (_, sample) =>
-          sample.summarySettings.get("single_sample_variantcalling") shouldBe Some(sampleCalling)
+          sample.summarySettings.get("single_sample_variantcalling") shouldBe None
           sample.summarySettings.get("use_indel_realigner") shouldBe Some(realign)
           sample.libraries.foreach {
             case (_, lib) =>
-              lib.summarySettings.get("library_variantcalling") shouldBe Some(libraryCalling)
+              lib.summarySettings.get("library_variantcalling") shouldBe None
               lib.summarySettings.get("use_indel_realigner") shouldBe None // Should not exist anymore
               lib.summarySettings.get("use_base_recalibration") shouldBe Some(
                 baseRecalibration && dbsnp)
           }
       }
 
-      pipeline.functions.count(_.isInstanceOf[VcfStats]) shouldBe ((if (multisampleCalling) 2
-                                                                    else 0) +
-        (if (sampleCalling) numberSamples * 2 else 0) +
-        (if (libraryCalling) numberLibs * 2 else 0))
+      pipeline.functions.count(_.isInstanceOf[VcfStats]) shouldBe (if (multisampleCalling) 2
+                                                                   else 0)
     }
   }
 
   // remove temporary run directory all tests in the class have been run
-  @AfterClass def removeTempOutputDir() = {
+  @AfterClass def removeTempOutputDir(): Unit = {
     dirs.filter(_.exists()).foreach { dir =>
       try {
         FileUtils.deleteDirectory(dir)
@@ -175,20 +169,6 @@ class ShivaNoPrintReadsTest extends ShivaTestTrait {
   override def realignProvider = Array(x = false)
   override def usePrintReads = false
 }
-class ShivaLibraryCallingTest extends ShivaTestTrait {
-  override def sample1 = Array(true, false)
-  override def sample2 = Array(false, true)
-  override def realignProvider = Array(x = false)
-  override def baseRecalibrationProvider = Array(x = false)
-  override def libraryCalling = true
-}
-class ShivaSampleCallingTest extends ShivaTestTrait {
-  override def sample1 = Array(true, false)
-  override def sample2 = Array(false, true)
-  override def realignProvider = Array(x = false)
-  override def baseRecalibrationProvider = Array(x = false)
-  override def sampleCalling = true
-}
 class ShivaWithSvCallingTest extends ShivaTestTrait {
   override def sample1 = Array(x = true)
   override def sample2 = Array(x = false)