Remove back reference between VariantAnnotation and Variant

README.md

@@ -62,7 +62,6 @@ ADAM ACTIONS
 CONVERSION OPERATIONS
             vcf2adam : Convert a VCF file to the corresponding ADAM format
             adam2vcf : Convert an ADAM variant to the VCF ADAM format
-           anno2adam : Convert a annotation file (in VCF format) to the corresponding ADAM format
           fasta2adam : Converts a text FASTA sequence file into an ADAMNucleotideContig Parquet file which represents assembled sequences.
           adam2fasta : Convert ADAM nucleotide contig fragments to FASTA files
    transformFeatures : Convert a file with sequence features into corresponding ADAM format and vice versa

adam-apis/src/main/scala/org/bdgenomics/adam/apis/java/JavaADAMContext.scala

@@ -25,8 +25,7 @@ import org.bdgenomics.adam.rdd.fragment.FragmentRDD
 import org.bdgenomics.adam.rdd.read.AlignmentRecordRDD
 import org.bdgenomics.adam.rdd.variant.{
   GenotypeRDD,
-  VariantRDD,
-  VariantAnnotationRDD
+  VariantRDD
 }
 
 object JavaADAMContext {
@@ -84,16 +83,6 @@ class JavaADAMContext(val ac: ADAMContext) extends Serializable {
     ac.loadFeatures(filePath)
   }
 
-  /**
-   * Loads in variant annotations.
-   *
-   * @param filePath The path to load the file from.
-   * @return Returns a VariantAnnotationRDD.
-   */
-  def loadVariantAnnotations(filePath: java.lang.String): VariantAnnotationRDD = {
-    ac.loadVariantAnnotations(filePath)
-  }
-
   /**
    * Loads in genotypes.
    *

adam-apis/src/test/scala/org/bdgenomics/adam/apis/java/JavaADAMContextSuite.scala

@@ -86,14 +86,4 @@ class JavaADAMContextSuite extends ADAMFunSuite {
 
     assert(newRdd.jrdd.count() === 6)
   }
-
-  ignore("can read and write a small .vcf as annotations") {
-    val path = copyResource("small.vcf")
-    val aRdd = sc.loadVariantAnnotations(path)
-    assert(aRdd.jrdd.count() === 6)
-
-    val newRdd = JavaADAMAnnotationConduit.conduit(aRdd, sc)
-
-    assert(newRdd.jrdd.count() === 6)
-  }
 }

adam-cli/src/main/scala/org/bdgenomics/adam/cli/ADAMMain.scala

@@ -46,7 +46,6 @@ object ADAMMain {
         List(
           Vcf2ADAM,
           ADAM2Vcf,
-          VcfAnnotation2ADAM,
           Fasta2ADAM,
           ADAM2Fasta,
           ADAM2Fastq,

adam-core/src/main/scala/org/bdgenomics/adam/converters/VariantContextConverter.scala

@@ -322,10 +322,7 @@ private[adam] class VariantContextConverter(
       builder.setFiltersPassed(!vc.isFiltered)
     }
     if (vc.isFiltered) {
-      builder.setFiltersFailed(new java.util.ArrayList(vc.getFilters));
-    }
-    if (vc.getAttributeAsBoolean("SOMATIC", false)) {
-      builder.setSomatic(true)
+      builder.setFiltersFailed(new java.util.ArrayList(vc.getFilters))
     }
     builder.build
   }
@@ -1234,12 +1231,6 @@ private[adam] class VariantContextConverter(
       case (true, true)   => vcb.passFilters
     }
 
-    val somatic: java.lang.Boolean = Option(variant.getSomatic).getOrElse(false)
-    if (somatic) {
-      vcb.attribute("SOMATIC", true)
-    }
-
-    // attach genotypes
     try {
       Some(vcb.genotypes(vc.genotypes.map(g => bdgConvFn(g)))
         .make)

adam-core/src/main/scala/org/bdgenomics/adam/models/ReferenceRegion.scala

@@ -169,16 +169,6 @@ object ReferenceRegion {
     ReferenceRegion(variant.getContigName, variant.getStart, variant.getEnd)
   }
 
-  /**
-   * Builds a reference region from a variant annotation.
-   *
-   * @param annotation VariantAnnotation to extract region from.
-   * @return The site where the variant for the specified variant annotation covers.
-   */
-  def apply(annotation: VariantAnnotation): ReferenceRegion = {
-    ReferenceRegion(annotation.getVariant)
-  }
-
   private def checkRead(record: AlignmentRecord) {
     require(record.getReadMapped,
       "Cannot build reference region for unmapped read %s.".format(record))

adam-core/src/main/scala/org/bdgenomics/adam/projections/VariantAnnotationField.scala

@@ -24,5 +24,5 @@ import org.bdgenomics.formats.avro.VariantAnnotation
  */
 object VariantAnnotationField extends FieldEnumeration(VariantAnnotation.SCHEMA$) {
 
-  val variant, ancestralAllele, alleleCount, readDepth, forwardReadDepth, reverseReadDepth, alleleFrequency, cigar, dbSnp, hapMap2, hapMap3, validated, thousandGenomes, transcriptEffects, attributes = SchemaValue
+  val ancestralAllele, alleleCount, readDepth, forwardReadDepth, reverseReadDepth, alleleFrequency, cigar, dbSnp, hapMap2, hapMap3, validated, thousandGenomes, somatic, transcriptEffects, attributes = SchemaValue
 }

adam-core/src/main/scala/org/bdgenomics/adam/projections/VariantField.scala

@@ -24,5 +24,5 @@ import org.bdgenomics.formats.avro.Variant
  */
 object VariantField extends FieldEnumeration(Variant.SCHEMA$) {
 
-  val contigName, start, end, names, referenceAllele, alternateAllele, filtersApplied, filtersPassed, filtersFailed, somatic = SchemaValue
+  val contigName, start, end, names, referenceAllele, alternateAllele, filtersApplied, filtersPassed, filtersFailed = SchemaValue
 }

adam-core/src/main/scala/org/bdgenomics/adam/rdd/ADAMContext.scala

@@ -1267,64 +1267,6 @@ class ADAMContext(@transient val sc: SparkContext) extends Serializable with Log
     FragmentRDD(rdd, sd, rgd)
   }
 
-  /**
-   * Loads variant annotations stored in VCF format.
-   *
-   * @param filePath The path to the VCF file(s) to load annotations from.
-   * @return Returns VariantAnnotationRDD.
-   */
-  def loadVcfAnnotations(
-    filePath: String): VariantAnnotationRDD = {
-    loadVcf(filePath).toVariantAnnotationRDD
-  }
-
-  /**
-   * Loads VariantAnnotations stored in Parquet, with metadata.
-   *
-   * @param filePath The path to load files from.
-   * @param predicate An optional predicate to push down into the file.
-   * @param projection An optional projection to use for reading.
-   * @return Returns VariantAnnotationRDD.
-   */
-  def loadParquetVariantAnnotations(
-    filePath: String,
-    predicate: Option[FilterPredicate] = None,
-    projection: Option[Schema] = None): VariantAnnotationRDD = {
-
-    // load header lines
-    val headers = loadHeaderLines(filePath)
-
-    val sd = loadAvroSequences(filePath)
-    val rdd = loadParquet[VariantAnnotation](filePath, predicate, projection)
-
-    VariantAnnotationRDD(rdd, sd, headers)
-  }
-
-  /**
-   * Loads VariantAnnotations into an RDD, and automatically detects
-   * the underlying storage format.
-   *
-   * Can load variant annotations from either Parquet or VCF.
-   *
-   * @see loadVcfAnnotations
-   * @see loadParquetVariantAnnotations
-   *
-   * @param filePath The path to load files from.
-   * @param projection An optional projection to use for reading.
-   * @return Returns VariantAnnotationRDD.
-   */
-  def loadVariantAnnotations(
-    filePath: String,
-    projection: Option[Schema] = None): VariantAnnotationRDD = {
-    if (filePath.endsWith(".vcf")) {
-      log.info(s"Loading $filePath as VCF, and converting to variant annotations. Projection is ignored.")
-      loadVcfAnnotations(filePath)
-    } else {
-      log.info(s"Loading $filePath as Parquet containing VariantAnnotations.")
-      loadParquetVariantAnnotations(filePath, None, projection)
-    }
-  }
-
   /**
    * Loads Features from a file, autodetecting the file type.
    *

adam-core/src/main/scala/org/bdgenomics/adam/rdd/variant/VariantContextRDD.scala

@@ -60,29 +60,6 @@ case class VariantContextRDD(rdd: RDD[VariantContext],
                              @transient headerLines: Seq[VCFHeaderLine] = SupportedHeaderLines.allHeaderLines) extends MultisampleGenomicRDD[VariantContext, VariantContextRDD]
     with Logging {
 
-  /**
-   * Left outer join database variant annotations.
-   *
-   * @param ann Annotation RDD to join against.
-   * @return Returns a VariantContextRDD where annotations have been filled in.
-   */
-  def joinVariantAnnotations(ann: VariantAnnotationRDD): VariantContextRDD = {
-    replaceRdd(rdd.keyBy(_.variant)
-      .leftOuterJoin(ann.rdd.keyBy(dba => RichVariant(dba.getVariant)))
-      .values
-      .map(kv => VariantContext(kv._1, kv._2)))
-  }
-
-  /**
-   * @return Returns a VariantAnnotationRDD containing the variant
-   *   annotations attached to this VariantContextRDD.
-   */
-  def toVariantAnnotationRDD: VariantAnnotationRDD = {
-    VariantAnnotationRDD(rdd.flatMap(_.annotations),
-      sequences,
-      headerLines)
-  }
-
   /**
    * @return Returns a GenotypeRDD containing the Genotypes in this RDD.
    */