[AVOCADO] Update ADAM dependency to 0.26.0.

avocado-cli/src/main/scala/org/bdgenomics/avocado/cli/BiallelicGenotyper.scala

@@ -216,11 +216,11 @@ class BiallelicGenotyper(
 
     // load reads
     val projection = Some(Filter(AlignmentRecordField.attributes,
-      AlignmentRecordField.origQual,
-      AlignmentRecordField.recordGroupName))
+      AlignmentRecordField.originalQuality,
+      AlignmentRecordField.readGroupId))
     val reads = sc.loadAlignments(args.inputPath,
       optProjection = projection)
-    val samples = reads.recordGroups.recordGroups.map(_.sample).toSet
+    val samples = reads.readGroups.readGroups.map(_.sampleId).toSet
     require(samples.nonEmpty,
       "Didn't see any samples attached to input. Did you forget to add read groups?")
     require(samples.size <= 1,

avocado-cli/src/main/scala/org/bdgenomics/avocado/cli/MergeDiscovered.scala

@@ -56,7 +56,7 @@ class MergeDiscovered(
     sc.loadVariants(args.inputPath)
       .transformDataset(_.dropDuplicates("start",
         "end",
-        "contigName",
+        "referenceName",
         "referenceAllele",
         "alternateAllele"))
       .saveAsParquet(args.outputPath)

avocado-cli/src/main/scala/org/bdgenomics/avocado/cli/Reassemble.scala

@@ -20,7 +20,7 @@ package org.bdgenomics.avocado.cli
 import org.apache.spark.SparkContext
 import org.bdgenomics.adam.rdd.ADAMContext._
 import org.bdgenomics.adam.rdd.ADAMSaveAnyArgs
-import org.bdgenomics.adam.rdd.read.{ AlignmentRecordRDD, MDTagging }
+import org.bdgenomics.adam.rdd.read.{ AlignmentRecordDataset, MDTagging }
 import org.bdgenomics.avocado.realigner.Realigner
 import org.bdgenomics.utils.cli._
 import org.kohsuke.args4j.{ Argument, Option => Args4jOption }

avocado-cli/src/main/scala/org/bdgenomics/avocado/cli/TrioGenotyper.scala

@@ -21,8 +21,8 @@ import org.apache.spark.SparkContext
 import org.bdgenomics.adam.projections.{ AlignmentRecordField, Filter }
 import org.bdgenomics.adam.rdd.ADAMContext._
 import org.bdgenomics.adam.rdd.ADAMSaveAnyArgs
-import org.bdgenomics.adam.rdd.read.AlignmentRecordRDD
-import org.bdgenomics.adam.rdd.variant.GenotypeRDD
+import org.bdgenomics.adam.rdd.read.AlignmentRecordDataset
+import org.bdgenomics.adam.rdd.variant.GenotypeDataset
 import org.bdgenomics.avocado.genotyping.{
   BiallelicGenotyper => Biallelic,
   DiscoverVariants => Discover,
@@ -198,8 +198,8 @@ class TrioGenotyper(
 
     // load reads
     val projection = Some(Filter(AlignmentRecordField.attributes,
-      AlignmentRecordField.origQual,
-      AlignmentRecordField.recordGroupName))
+      AlignmentRecordField.originalQuality,
+      AlignmentRecordField.readGroupId))
     val firstParentReads = sc.loadAlignments(args.firstParentPath,
       optProjection = projection)
     val secondParentReads = sc.loadAlignments(args.secondParentPath,
@@ -250,7 +250,7 @@ class TrioGenotyper(
       copyNumber,
       false)
 
-    val genotypes = GenotypeRDD(sc.union(firstParentGenotypes.rdd,
+    val genotypes = GenotypeDataset(sc.union(firstParentGenotypes.rdd,
       secondParentGenotypes.rdd,
       childGenotypes.rdd),
       variants.sequences,

avocado-core/src/main/scala/org/bdgenomics/avocado/genotyping/BiallelicGenotyper.scala

@@ -25,10 +25,10 @@ import org.apache.spark.sql.SQLContext
 import org.apache.spark.sql.types.IntegerType
 import org.bdgenomics.adam.models.ReferenceRegion
 import org.bdgenomics.adam.rdd.GenomeBins
-import org.bdgenomics.adam.rdd.read.AlignmentRecordRDD
+import org.bdgenomics.adam.rdd.read.AlignmentRecordDataset
 import org.bdgenomics.adam.rdd.variant.{
-  GenotypeRDD,
-  VariantRDD
+  GenotypeDataset,
+  VariantDataset
 }
 import org.bdgenomics.adam.util.PhredUtils
 import org.bdgenomics.avocado.Timers._
@@ -85,21 +85,21 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
    * @param maxMapQ The highest mapping quality to allow.
    * @return Returns genotype calls.
    */
-  def call(reads: AlignmentRecordRDD,
-           variants: VariantRDD,
+  def call(reads: AlignmentRecordDataset,
+           variants: VariantDataset,
            copyNumber: CopyNumberMap,
            scoreAllSites: Boolean,
            optDesiredPartitionCount: Option[Int] = None,
            optDesiredPartitionSize: Option[Int] = None,
            optDesiredMaxCoverage: Option[Int] = None,
            maxQuality: Int = 93,
-           maxMapQ: Int = 93): GenotypeRDD = CallGenotypes.time {
+           maxMapQ: Int = 93): GenotypeDataset = CallGenotypes.time {
 
     // validate metadata
     require(variants.sequences.isCompatibleWith(reads.sequences),
       "Variant sequence dictionary (%s) is not compatible with read dictionary (%s).".format(
         variants.sequences, reads.sequences))
-    val samples = reads.recordGroups.recordGroups.map(_.sample).toSet
+    val samples = reads.readGroups.readGroups.map(_.sampleId).toSet
     require(samples.size == 1,
       "Currently, we only support a single sample. Saw: %s.".format(
         samples.mkString(", ")))
@@ -124,11 +124,11 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
     val genotypeRdd = observationsToGenotypes(observationRdd,
       samples.head)
 
-    GenotypeRDD(genotypeRdd,
+    GenotypeDataset(genotypeRdd,
       variants.sequences,
       samples.map(s => {
         Sample.newBuilder()
-          .setSampleId(s)
+          .setId(s)
           .setName(s)
           .build()
       }).toSeq, org.bdgenomics.adam.converters.DefaultHeaderLines.allHeaderLines)
@@ -153,7 +153,7 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
    * @param maxMapQ The highest mapping quality to allow.
    * @return Returns genotype calls.
    */
-  def discoverAndCall(reads: AlignmentRecordRDD,
+  def discoverAndCall(reads: AlignmentRecordDataset,
                       copyNumber: CopyNumberMap,
                       scoreAllSites: Boolean,
                       optDesiredPartitionCount: Option[Int] = None,
@@ -162,7 +162,7 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
                       optDesiredPartitionSize: Option[Int] = None,
                       optDesiredMaxCoverage: Option[Int] = None,
                       maxQuality: Int = 93,
-                      maxMapQ: Int = 93): GenotypeRDD = {
+                      maxMapQ: Int = 93): GenotypeDataset = {
 
     // get rdd storage level and warn if not persisted
     val readSl = reads.rdd.getStorageLevel
@@ -442,7 +442,7 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
 
     // flatten schema
     val flatFields = Seq(
-      observationsDf("_1.contigName").as("contigName"),
+      observationsDf("_1.referenceName").as("referenceName"),
       observationsDf("_1.start").as("start"),
       observationsDf("_1.referenceAllele").as("referenceAllele"),
       observationsDf("_1.alternateAllele").as("alternateAllele"),
@@ -493,14 +493,14 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
         sum("totalCoverage").as("totalCoverage"),
         first("isRef").as("isRef"),
         first("copyNumber").as("copyNumber"))
-    val aggregatedObservationsDf = joinedObservationsDf.groupBy("contigName",
+    val aggregatedObservationsDf = joinedObservationsDf.groupBy("referenceName",
       "start",
       "referenceAllele",
       "alternateAllele")
       .agg(aggCols.head, aggCols.tail: _*)
 
     // re-nest the output
-    val firstField = struct(aggregatedObservationsDf("contigName"),
+    val firstField = struct(aggregatedObservationsDf("referenceName"),
       aggregatedObservationsDf("start"),
       aggregatedObservationsDf("referenceAllele"),
       aggregatedObservationsDf("alternateAllele"))
@@ -733,7 +733,7 @@ private[avocado] object BiallelicGenotyper extends Serializable with Logging {
       .setVariantCallingAnnotations(vcAnnotations)
       .setStart(v.getStart)
       .setEnd(v.getEnd)
-      .setContigName(v.getContigName)
+      .setReferenceName(v.getReferenceName)
       .setSampleId(sample)
       .setStrandBiasComponents(sbComponents
         .map(i => i: java.lang.Integer))

avocado-core/src/main/scala/org/bdgenomics/avocado/genotyping/DiscoverVariants.scala

@@ -19,8 +19,8 @@ package org.bdgenomics.avocado.genotyping
 
 import org.apache.spark.rdd.RDD
 import org.apache.spark.sql.SQLContext
-import org.bdgenomics.adam.rdd.read.AlignmentRecordRDD
-import org.bdgenomics.adam.rdd.variant.VariantRDD
+import org.bdgenomics.adam.rdd.read.AlignmentRecordDataset
+import org.bdgenomics.adam.rdd.variant.VariantDataset
 import org.bdgenomics.avocado.Timers._
 import org.bdgenomics.avocado.models.{
   Clipped,
@@ -41,19 +41,19 @@ import scala.annotation.tailrec
 object DiscoverVariants extends Serializable with Logging {
 
   /**
-   * Discovers all variants in an RDD of reads.
+   * Discovers all variants in an dataset of reads.
    *
-   * @param aRdd RDD of reads.
+   * @param aRdd Dataset of reads.
    * @param optPhredThreshold An optional threshold that discards all variants
    *   not supported by bases of at least a given phred score.
-   * @return Returns an RDD of variants.
+   * @return Returns a dataset of variants.
    */
   private[avocado] def apply(
-    aRdd: AlignmentRecordRDD,
+    aRdd: AlignmentRecordDataset,
     optPhredThreshold: Option[Int] = None,
-    optMinObservations: Option[Int] = None): VariantRDD = DiscoveringVariants.time {
+    optMinObservations: Option[Int] = None): VariantDataset = DiscoveringVariants.time {
 
-    VariantRDD(variantsInRdd(aRdd.rdd,
+    VariantDataset(variantsInRdd(aRdd.rdd,
       optPhredThreshold = optPhredThreshold,
       optMinObservations = optMinObservations),
       aRdd.sequences,
@@ -87,7 +87,7 @@ object DiscoverVariants extends Serializable with Logging {
     val uniqueVariants = optMinObservations.fold({
       variantDs.distinct
     })(mo => {
-      variantDs.groupBy(variantDs("contigName"),
+      variantDs.groupBy(variantDs("referenceName"),
         variantDs("start"),
         variantDs("referenceAllele"),
         variantDs("alternateAllele"))
@@ -132,8 +132,8 @@ object DiscoverVariants extends Serializable with Logging {
 
       // get the read sequence, contig, etc
       val sequence = read.getSequence
-      val qual = read.getQual
-      val contigName = read.getContigName
+      val qual = read.getQuality
+      val referenceName = read.getReferenceName
 
       // advance to the first alignment match
       @tailrec def fastForward(
@@ -198,7 +198,7 @@ object DiscoverVariants extends Serializable with Logging {
                 val newVars = (0 until length).flatMap(i => {
                   if (qual(i).toInt - 33 >= phredThreshold) {
                     Some(DiscoveredVariant(
-                      contigName,
+                      referenceName,
                       pos + i,
                       ref(i).toString,
                       sequence(idx + i).toString))
@@ -216,7 +216,7 @@ object DiscoverVariants extends Serializable with Logging {
               val insQuals = qual.substring(idx - 1, idx + length).map(_.toInt - 33).sum / length
               val newVar = if (insQuals >= phredThreshold) {
                 DiscoveredVariant(
-                  contigName,
+                  referenceName,
                   pos - 1,
                   lastRef,
                   sequence.substring(idx - 1, idx + length)) :: variants
@@ -230,7 +230,7 @@ object DiscoverVariants extends Serializable with Logging {
               val delLength = ref.size
               val newVar = if (qual(idx - 1).toInt - 33 >= phredThreshold) {
                 DiscoveredVariant(
-                  contigName,
+                  referenceName,
                   pos - 1,
                   lastRef + ref,
                   sequence.substring(idx - 1, idx)) :: variants

avocado-core/src/main/scala/org/bdgenomics/avocado/genotyping/DiscoveredVariant.scala

@@ -30,25 +30,25 @@ private[genotyping] object DiscoveredVariant {
    * @return Returns a case class-based representation of the variant.
    */
   def apply(variant: Variant): DiscoveredVariant = {
-    new DiscoveredVariant(variant.getContigName,
+    new DiscoveredVariant(variant.getReferenceName,
       variant.getStart.toInt,
       variant.getReferenceAllele,
       Some(variant.getAlternateAllele))
   }
 
   /**
-   * @param contigName The contig this variant is on.
+   * @param referenceName The contig this variant is on.
    * @param start The position this variant starts at.
    * @param referenceAllele The reference allele this variant varies from.
    * @param alternateAllele The substituted allele.
    * @return Returns a discovered variant with a defined alternate allele.
    */
   def apply(
-    contigName: String,
+    referenceName: String,
     start: Int,
     referenceAllele: String,
     alternateAllele: String): DiscoveredVariant = {
-    new DiscoveredVariant(contigName, start, referenceAllele, Some(alternateAllele))
+    new DiscoveredVariant(referenceName, start, referenceAllele, Some(alternateAllele))
   }
 
   /**
@@ -64,13 +64,13 @@ private[genotyping] object DiscoveredVariant {
 /**
  * A variant site and alleles.
  *
- * @param contigName The contig this variant is on.
+ * @param referenceName The reference this variant is on.
  * @param start The position this variant starts at.
  * @param referenceAllele The reference allele this variant varies from.
  * @param alternateAllele The substituted allele.
  */
 case class DiscoveredVariant(
-    contigName: String,
+    referenceName: String,
     start: Int,
     referenceAllele: String,
     alternateAllele: Option[String]) {
@@ -87,7 +87,7 @@ case class DiscoveredVariant(
    */
   def toVariant: Variant = {
     val builder = Variant.newBuilder
-      .setContigName(contigName)
+      .setReferenceName(referenceName)
       .setStart(start.toLong)
       .setEnd(end.toLong)
       .setReferenceAllele(referenceAllele)
@@ -100,10 +100,10 @@ case class DiscoveredVariant(
   }
 
   def overlaps(v: DiscoveredVariant): Boolean = {
-    contigName == v.contigName && start < v.end && end > v.start
+    referenceName == v.referenceName && start < v.end && end > v.start
   }
 
   def overlaps(rr: ReferenceRegion): Boolean = {
-    contigName == rr.referenceName && start < rr.end && end > rr.start
+    referenceName == rr.referenceName && start < rr.end && end > rr.start
   }
 }

avocado-core/src/main/scala/org/bdgenomics/avocado/genotyping/JointAnnotatorCaller.scala

@@ -21,8 +21,8 @@ import breeze.stats.distributions.Binomial
 import org.apache.spark.rdd.RDD
 import org.bdgenomics.adam.models.VariantContext
 import org.bdgenomics.adam.rdd.variant.{
-  GenotypeRDD,
-  VariantContextRDD
+  GenotypeDataset,
+  VariantContextDataset
 }
 import org.bdgenomics.adam.util.PhredUtils
 import org.bdgenomics.avocado.util.LogUtils
@@ -49,7 +49,7 @@ object JointAnnotatorCaller extends Serializable {
    * @param genotypes The genotypes to jointly process.
    * @return Returns a squared off and annotated set of variant contexts.
    */
-  def apply(genotypes: GenotypeRDD): VariantContextRDD = {
+  def apply(genotypes: GenotypeDataset): VariantContextDataset = {
     apply(genotypes.toVariantContexts)
   }
 
@@ -59,7 +59,7 @@ object JointAnnotatorCaller extends Serializable {
    * @param variantContexts The squared off sites to process.
    * @return Returns a squared off and annotated set of variant contexts.
    */
-  def apply(variantContexts: VariantContextRDD): VariantContextRDD = {
+  def apply(variantContexts: VariantContextDataset): VariantContextDataset = {
     variantContexts.transform(_.flatMap(annotateSite))
   }
 

avocado-core/src/main/scala/org/bdgenomics/avocado/genotyping/Observer.scala

@@ -52,11 +52,11 @@ private[genotyping] object Observer extends Serializable {
 
     // for convenience, get the sample name, mapping quality, sequence,
     // qualities, and the contig name
-    val sampleId = read.getRecordGroupSample
-    val contigName = read.getContigName
-    val mapQ = read.getMapq
+    val sampleId = read.getReadGroupSampleId
+    val referenceName = read.getReferenceName
+    val mapQ = read.getMappingQuality
     val readSequence = read.getSequence
-    val readQualities = read.getQual
+    val readQualities = read.getQuality
     val forwardStrand = !read.getReadNegativeStrand
 
     // map over the alignment operators and generate allelic observations
@@ -73,7 +73,7 @@ private[genotyping] object Observer extends Serializable {
         (0 until length).map(idx => {
 
           // the key is the (site, allele, sampleId)
-          val key = (ReferenceRegion(contigName, pos, pos + 1),
+          val key = (ReferenceRegion(referenceName, pos, pos + 1),
             readSequence(readIdx).toString,
             sampleId)
 
@@ -104,7 +104,7 @@ private[genotyping] object Observer extends Serializable {
 
         // the key is the (site, allele, sampleId)
         // insertions associate to the site to their left, hence the -1
-        val key = (ReferenceRegion(contigName, pos - 1, pos),
+        val key = (ReferenceRegion(referenceName, pos - 1, pos),
           bases,
           sampleId)
 
@@ -124,7 +124,7 @@ private[genotyping] object Observer extends Serializable {
 
         // the key is the (site, allele, sampleId)
         // deletions have an empty string for the allele
-        val key = (ReferenceRegion(contigName, oldPos, pos),
+        val key = (ReferenceRegion(referenceName, oldPos, pos),
           "",
           sampleId)