Add a tool to make IGV-compat. BEDPE files from AggregateSvPileup

src/main/scala/com/fulcrumgenomics/sv/cmdline/ClpGroups.scala

@@ -5,10 +5,16 @@ import com.fulcrumgenomics.sopt.cmdline.ClpGroup
 /** Groups for organizing command line programs for display. */
 object ClpGroups {
 
-  class _All extends ClpGroup {
-    override val name: String = "All tools"
-    override val description: String = "All tools."
+  class _BreakpointAndSv extends ClpGroup {
+    override val name: String = "Breakpoint and SV Tools"
+    override val description: String = "Primary tools for calling and transforming breakpoints and SVs."
   }
 
-  final val All = classOf[_All]
+  class _Utilities  extends ClpGroup {
+    override val name: String = "Utility Tools"
+    override val description: String = "Helper tools for working with breakpoint or SV data."
+  }
+
+  final val BreakpointAndSv = classOf[_BreakpointAndSv]
+  final val Utilities = classOf[_Utilities]
 }

src/main/scala/com/fulcrumgenomics/sv/tools/AggregateSvPileup.scala

@@ -15,7 +15,7 @@ import com.fulcrumgenomics.FgBioDef._
 import scala.collection.mutable
 
 
-@clp(group=ClpGroups.All, description=
+@clp(group=ClpGroups.BreakpointAndSv, description=
   """
     |Merges nearby pileups of reads supporting putative breakpoints.
     |

src/main/scala/com/fulcrumgenomics/sv/tools/AggregateSvPileupToBedPE.scala

@@ -0,0 +1,127 @@
+package com.fulcrumgenomics.sv.tools
+
+import com.fulcrumgenomics.commons.CommonsDef._
+import com.fulcrumgenomics.commons.io.Writer
+import com.fulcrumgenomics.commons.util.LazyLogging
+import com.fulcrumgenomics.sopt.{arg, clp}
+import com.fulcrumgenomics.sv.cmdline.{ClpGroups, SvTool}
+import com.fulcrumgenomics.sv.tools.AggregateSvPileupToBedPE.BedPE
+import com.fulcrumgenomics.sv.tools.AggregateSvPileupToBedPE.BedPE.BedPEWriter
+import com.fulcrumgenomics.util.{Io, Metric}
+import htsjdk.tribble.annotation.Strand
+
+import java.io.BufferedWriter
+
+
+@clp(group=ClpGroups.Utilities, description= "Convert the output of AggregateSvPileup to BEDPE.")
+class AggregateSvPileupToBedPE(
+  @arg(flag='i', doc="Input text file of aggregate pileups generated by AggregateSvPileup") input: FilePath,
+  @arg(flag='o', doc="Output text file of the aggregate pileups in BEDPE format.") output: FilePath,
+) extends SvTool with LazyLogging {
+
+  Io.assertReadable(input)
+  Io.assertCanWriteFile(output)
+
+  override def execute(): Unit = {
+    val reader = Metric.iterator[AggregatedBreakpointPileup](input)
+    val writer = BedPEWriter(output)
+
+    reader.map(BedPE.apply).foreach(writer.write)
+
+    writer.close()
+  }
+}
+
+/** Companion object for [[AggregateSvPileupToBedPE]]. */
+object AggregateSvPileupToBedPE {
+
+  /** The IGV-supported BEDPE file extension. */
+  val BedPEExtension: String = ".bedpe"
+
+  /** A simple BEDPE record as defined by `bedtools`:
+    *
+    * - https://bedtools.readthedocs.io/en/latest/content/general-usage.html#bedpe-format)
+    *
+    * Future compatibility could be implemented for supporting 10x flavored BEDPE files:
+    *
+    * - https://github.com/igvteam/igv/wiki/BedPE-Support
+    *
+    * Note that the field `score` is allowed to be a string per bedtools!
+    */
+  case class BedPE(
+    chrom1: String,
+    start1: Int,
+    end1: Int,
+    chrom2: String,
+    start2: Int,
+    end2: Int,
+    name: String,
+    score: String,
+    strand1: Strand,
+    strand2: Strand,
+  ) extends Metric
+
+  /** Companion object for [[BedPE]]. */
+  object BedPE {
+
+    /** Build a [[BedPE]] record from an [[AggregatedBreakpointPileup]]. */
+    def apply(pileup: AggregatedBreakpointPileup): BedPE = {
+      new BedPE(
+        chrom1  = pileup.left_contig,
+        start1  = pileup.left_min_pos,
+        end1    = pileup.left_max_pos + 1,
+        chrom2  = pileup.right_contig,
+        start2  = pileup.right_min_pos,
+        end2    = pileup.right_max_pos + 1,
+        name    = pileup.id,
+        score   = pileup.total.toString,
+        strand1 = Strand.decode(pileup.left_strand),
+        strand2 = Strand.decode(pileup.right_strand),
+      )
+    }
+
+    /** A writer class for writing [[BedPE]] records. */
+    class BedPEWriter(val out: BufferedWriter) extends Writer[BedPE] {
+
+      /** Write a [[BedPE]] record to the underlying writer. */
+      override def write(record: BedPE): Unit = {
+        out.write(record.chrom1)
+        out.write('\t')
+        out.write(Integer.toString(record.start1))
+        out.write('\t')
+        out.write(Integer.toString(record.end1))
+        out.write('\t')
+        out.write(record.chrom2)
+        out.write('\t')
+        out.write(Integer.toString(record.start2))
+        out.write('\t')
+        out.write(Integer.toString(record.end2))
+        out.write('\t')
+        out.write(record.name)
+        out.write('\t')
+        out.write(record.score)
+        out.write('\t')
+        out.write(record.strand1.toString)
+        out.write('\t')
+        out.write(record.strand2.toString)
+        out.newLine()
+      }
+
+      /** Closes the underlying writer. */
+      override def close(): Unit = out.close()
+    }
+
+    /** Companion object to [[BedPEWriter]]. */
+    object BedPEWriter {
+
+      /** Constructs a [[BedPEWriter]] that will write to the provided path. */
+      def apply(path: PathToIntervals): BedPEWriter = apply(Io.toWriter(path))
+
+      /** Constructs a [[BedPEWriter]] from a [[java.io.Writer]]. */
+      def apply(writer: java.io.Writer): BedPEWriter = writer match {
+        case bw: BufferedWriter => new BedPEWriter(bw)
+        case w                  => new BedPEWriter(new BufferedWriter(w))
+      }
+    }
+  }
+}

src/main/scala/com/fulcrumgenomics/sv/tools/FilterAndMerge.scala

@@ -10,7 +10,7 @@ import com.fulcrumgenomics.util.{Io, Metric}
 
 import scala.collection.immutable.IndexedSeq
 
-@clp(group=ClpGroups.All, description=
+@clp(group=ClpGroups.BreakpointAndSv, description=
   """
     |Filters and merges SVPileup output.
   """)

src/main/scala/com/fulcrumgenomics/sv/tools/SvPileup.scala

@@ -37,7 +37,7 @@ object TargetBedRequirement extends FgBioEnum[TargetBedRequirement] {
 }
 
 
-@clp(group=ClpGroups.All, description=
+@clp(group=ClpGroups.BreakpointAndSv, description=
   """
     |Collates a pileup of putative structural variant supporting reads.
     |

src/test/scala/com/fulcrumgenomics/sv/tools/AggregateSvPileupToBedPETest.scala

@@ -0,0 +1,117 @@
+package com.fulcrumgenomics.sv.tools
+
+import com.fulcrumgenomics.commons.io.Io
+import com.fulcrumgenomics.commons.util.DelimitedDataParser
+import com.fulcrumgenomics.sv.UnitSpec
+import com.fulcrumgenomics.sv.tools.AggregateSvPileupToBedPE.BedPE.BedPEWriter
+import com.fulcrumgenomics.sv.tools.AggregateSvPileupToBedPE.{BedPE, BedPEExtension}
+import com.fulcrumgenomics.util.Metric
+import htsjdk.tribble.annotation.Strand
+
+
+/** Unit tests for [[AggregateSvPileupToBedPE]]. */
+class AggregateSvPileupToBedPETest extends UnitSpec {
+
+  /** A test aggregate breakpoint pileup. */
+  private val test_aggregate_breakpoint_pileup = AggregatedBreakpointPileup(
+    id             = "112",
+    category       = "Inter-contig rearrangement",
+    left_contig    = "chr1",
+    left_min_pos   = 100,
+    left_max_pos   = 100,
+    left_strand    = '+',
+    right_contig   = "chr3",
+    right_min_pos  = 200,
+    right_max_pos  = 200,
+    right_strand   = '-',
+    split_reads    = 1,
+    read_pairs     = 1,
+    total          = 2,
+    left_pileups   = PositionList(100),
+    right_pileups  = PositionList(200),
+  )
+
+  /** A companion test BEDPE record. */
+  private val test_bed_pe = BedPE(
+    chrom1  = "chr1",
+    start1  = 100,
+    end1    = 101,
+    chrom2  = "chr3",
+    start2  = 200,
+    end2    = 201,
+    name    = "112",
+    score   = "2",
+    strand1 = Strand.POSITIVE,
+    strand2 = Strand.NEGATIVE,
+  )
+
+  "AggregateSvPileupToBedPE.BedPE" should "be instantiated from an AggregateBreakpointPileup" in {
+    BedPE(test_aggregate_breakpoint_pileup) shouldBe test_bed_pe
+  }
+
+  "AggregateSvPileupToBedPE.BedPEWriter" should "write a BedPE record" in {
+    val record = new BedPE(
+      chrom1  = "chr1",
+      start1  = 100,
+      end1    = 101,
+      chrom2  = "chr3",
+      start2  = 200,
+      end2    = 201,
+      name    = "112",
+      score   = "2",
+      strand1 = Strand.POSITIVE,
+      strand2 = Strand.NEGATIVE,
+    )
+
+    val expected = Seq(
+      record.chrom1,
+      Integer.toString(record.start1),
+      Integer.toString(record.end1),
+      record.chrom2,
+      Integer.toString(record.start2),
+      Integer.toString(record.end2),
+      record.name,
+      record.score,
+      record.strand1.toString,
+      record.strand2.toString,
+    ).toIndexedSeq
+
+    val output = Io.makeTempFile(this.getClass.getSimpleName, BedPEExtension)
+    val writer = BedPEWriter(output)
+    writer.write(record)
+    writer.close()
+
+    val fields: Seq[String] = classOf[BedPE].getDeclaredFields.map(_.getName).toIndexedSeq
+    val records = DelimitedDataParser(output, delimiter = '\t', header = fields).toSeq
+    records.length shouldBe 1
+    val actual = fields.map(field => records.head.get[String](field).value)
+    actual should contain theSameElementsInOrderAs expected
+  }
+
+  "AggregateSvPileupToBedPE" should "convert an AggregateSvPileup output to a BEDPE file" in {
+    val expected = Seq(
+      test_bed_pe.chrom1,
+      Integer.toString(test_bed_pe.start1),
+      Integer.toString(test_bed_pe.end1),
+      test_bed_pe.chrom2,
+      Integer.toString(test_bed_pe.start2),
+      Integer.toString(test_bed_pe.end2),
+      test_bed_pe.name,
+      test_bed_pe.score,
+      test_bed_pe.strand1.toString,
+      test_bed_pe.strand2.toString,
+    ).toIndexedSeq
+
+    val input  = Io.makeTempFile(this.getClass.getSimpleName, ".txt")
+    val output = Io.makeTempFile(this.getClass.getSimpleName, BedPEExtension)
+    Metric.write[AggregatedBreakpointPileup](input, test_aggregate_breakpoint_pileup)
+
+    new AggregateSvPileupToBedPE(input = input, output = output).execute()
+
+    val fields: Seq[String] = classOf[BedPE].getDeclaredFields.map(_.getName).toIndexedSeq
+    val records = DelimitedDataParser(output, delimiter = '\t', header = fields).toSeq
+    records.length shouldBe 1
+    val actual = fields.map(field => records.head.get[String](field).value)
+    actual should contain theSameElementsInOrderAs expected
+  }
+}