GCCounter in Scala inspired by http://saml.rilspace.org/moar-languagez-gc-content-in-python-d-fpc-c-and-c. This is no match performance wise for the C/D/C++ etc solutions, but I think that it makes a nice point of showing how Scalas parallel collections can be used to with very little extra effort parallelize operations. Test file available for …

GCCounter.scala

import scala.io.Source
import java.io.File

object GCCounter extends App {

  val file = new File("Homo_sapiens.GRCh37.67.dna_rm.chromosome.Y.fa")

  // The actual GC counting function
  def countGCOnLine(line: String): (Long, Long) = {
    if (line.startsWith(">"))
      (0, 0)
    else {
      val at = line.count(c => c == 'A' || c == 'T')
      val gc = line.count(c => c == 'C' || c == 'G')
      (at + gc, gc)
    }
  }

  // Read 1 kb at the time
  val chunkSize = 1024 * 1
  val iterator = Source.fromFile(file, chunkSize).getLines.grouped(chunkSize)

  // Move through the iterator and accumulate the total gc over each chunck
  val (gc, total) = iterator.foldLeft[(Long, Long)]((0, 0))((accumulator, chunck) => {

    // Map/Reduce the gc over each line in the chunk in parallel.
    // The magic very simple parallelism magic going on here is in the 'par'
    // keyword which converts the chunck collection to a parallel collection
    // one which the map operations will automatically be parallelized.
    val (gc, total) = chunck.par.map(line => countGCOnLine(line)).
      reduce((x, y) => (x._1 + y._1, x._2 + y._2))

    (accumulator._1 + total, accumulator._2 + gc)
  })

  println("% GC: " + (gc.toFloat / total) * 100)

}

    

HowToCompileAndRun.sh

scalac -optimize GCCounter.scala
time scala GCCounter