Merge pull request #590 from fnothaft/remove-pileups

[ADAM-589] Removed pileup code.

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

@@ -53,8 +53,6 @@ object ADAMMain extends Logging {
  VcfAnnotation2ADAM,
  ADAM2Vcf,
  Fasta2ADAM,
- Reads2Ref,
- MpileupCommand,
  Features2ADAM,
  WigFix2Bed
  )

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

@@ -245,17 +245,6 @@ object ReferencePosition {
  None
  }
  }
-
- /**
- * Generates a reference position from a pileup base. Pileups are mapped by definition, so no
- * option wrapper is required.
- *
- * @param pileup A single pileup base.
- * @return The reference position of this pileup.
- */
- def apply(pileup: Pileup): ReferencePosition = {
- new ReferencePosition(pileup.getContig.getContigName, pileup.getPosition)
- }
 }
 
 case class ReferencePosition(referenceName: String, pos: Long) extends Ordered[ReferencePosition] {

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

@@ -35,7 +35,6 @@ import org.bdgenomics.adam.predicates.ADAMPredicate
 import org.bdgenomics.adam.projections.{ AlignmentRecordField, NucleotideContigFragmentField, Projection }
 import org.bdgenomics.adam.rdd.contig.NucleotideContigFragmentRDDFunctions
 import org.bdgenomics.adam.rdd.features._
-import org.bdgenomics.adam.rdd.pileup.{ PileupRDDFunctions, RodRDDFunctions }
 import org.bdgenomics.adam.rdd.read.AlignmentRecordRDDFunctions
 import org.bdgenomics.adam.rdd.variation._
 import org.bdgenomics.adam.rich.RichAlignmentRecord
@@ -62,12 +61,6 @@ object ADAMContext {
  // Add methods specific to Read RDDs
  implicit def rddToADAMRecordRDD(rdd: RDD[AlignmentRecord]) = new AlignmentRecordRDDFunctions(rdd)
 
- // Add methods specific to the Pileup RDDs
- implicit def rddToPileupRDD(rdd: RDD[Pileup]) = new PileupRDDFunctions(rdd)
-
- // Add methods specific to the Rod RDDs
- implicit def rddToRodRDD(rdd: RDD[Rod]) = new RodRDDFunctions(rdd)
-
  // Add methods specific to the ADAMNucleotideContig RDDs
  implicit def rddToContigFragmentRDD(rdd: RDD[NucleotideContigFragment]) = new NucleotideContigFragmentRDDFunctions(rdd)
 

adam-core/src/main/scala/org/bdgenomics/adam/rdd/read/AlignmentRecordRDDFunctions.scala

@@ -332,72 +332,6 @@ class AlignmentRecordRDDFunctions(rdd: RDD[AlignmentRecord])
  SingleReadBucket(rdd)
  }
 
- /**
- * Groups all reads by reference position and returns a non-aggregated pileup RDD.
- *
- * @param secondaryAlignments Creates pileups for non-primary aligned reads. Default is false.
- * @return Pileup without aggregation
- */
- def adamRecords2Pileup(secondaryAlignments: Boolean = false): RDD[Pileup] = {
- val helper = new Reads2PileupProcessor(secondaryAlignments)
- helper.process(rdd)
- }
-
- /**
- * Groups all reads by reference position, with all reference position bases grouped
- * into a rod.
- *
- * @param bucketSize Size in basepairs of buckets. Larger buckets take more time per
- * bucket to convert, but have lower skew. Default is 1000.
- * @param secondaryAlignments Creates rods for non-primary aligned reads. Default is false.
- * @return RDD of Rods.
- */
- def adamRecords2Rods(bucketSize: Int = 1000,
- secondaryAlignments: Boolean = false): RDD[Rod] = {
-
- /**
- * Maps a read to one or two buckets. A read maps to a single bucket if both
- * it's start and end are in a single bucket.
- *
- * @param r Read to map.
- * @return List containing one or two mapping key/value pairs.
- */
- def mapToBucket(r: AlignmentRecord): List[(ReferencePosition, AlignmentRecord)] = {
- val s = r.getStart / bucketSize
- val e = r.getEnd / bucketSize
- val name = r.getContig.getContigName
-
- if (s == e) {
- List((new ReferencePosition(name, s), r))
- } else {
- List((new ReferencePosition(name, s), r), (new ReferencePosition(name, e), r))
- }
- }
-
- val bucketedReads = rdd.filter(_.getStart != null)
- .flatMap(mapToBucket)
- .groupByKey()
-
- val pp = new Reads2PileupProcessor(secondaryAlignments)
-
- /**
- * Converts all reads in a bucket into rods.
- *
- * @param bucket Tuple of (bucket number, reads in bucket).
- * @return A sequence containing the rods in this bucket.
- */
- def bucketedReadsToRods(bucket: (ReferencePosition, Iterable[AlignmentRecord])): Iterable[Rod] = {
- val (_, bucketReads) = bucket
-
- bucketReads.flatMap(pp.readToPileups)
- .groupBy(ReferencePosition(_))
- .toList
- .map(g => Rod(g._1, g._2.toList)).toSeq
- }
-
- bucketedReads.flatMap(bucketedReadsToRods)
- }
-
  /**
  * Converts a set of records into an RDD containing the pairs of all unique tagStrings
  * within the records, along with the count (number of records) which have that particular