The Computer Language
Benchmarks Game

k-nucleotide F# .NET Core program

source code

(* The Computer Language Benchmarks Game
 * http://benchmarksgame.alioth.debian.org/
 *
 * contributed by Jimmy Tang
 *)
open System
open System.IO
open System.Collections.Generic

// make our hashtable using System.Collections.Generic.Dictionary
let maketable (dna:string) (length:int) =
   let d = new Dictionary<_,_>()
   for start in 0..(dna.Length - length) do
      let substr = dna.Substring(start, length)
      let x = ref (ref 0)
      if d.TryGetValue(substr, x) then x.Value := ! !x + 1
      else d.[substr] <- ref 1
   d

// frequency for all substrings of a given length
let frequencies (dna:string) (length:int) = [
   let d = maketable dna length
   let total = d.Values |> Seq.map (!) |> Seq.sum
   yield! [ for pair in d ->
             pair.Key.ToUpper(), (float(pair.Value.Value) * 100.0 /float(total))]
         |> List.sortBy (snd >> (~-))
         |> List.map (fun (s,c) -> sprintf "%s %.3f" s c)
   yield ""
]

// frequency of occurrence for a particular substring
let countSubstring dna (substring:string) = [
   let d = maketable dna (substring.Length)
   yield (sprintf "%d\t%s"
      (if d.ContainsKey(substring) then !d.[substring] else 0)
      (substring.ToUpper()))
]

let input = Console.In
let dna = seq { while true do yield input.ReadLine() }
        |> Seq.takeWhile (fun x -> x <> null)
        |> Seq.skipWhile (fun x -> not(x.StartsWith(">THREE")))
        |> Seq.skip 1
        |> String.concat ""

[for len in [1;2] -> async { return frequencies dna len }] @
[for str in ["ggt";"ggta";"ggtatt";"ggtattttaatt";"ggtattttaatttatagt"]
 -> async { return countSubstring dna str }]
|> List.rev
|> Async.Parallel
|> Async.RunSynchronously
|> Array.rev
|> Seq.concat
|> Seq.iter (printfn "%s")
    

notes, command-line, and program output

NOTES:
64-bit Ubuntu quad core
F# 4.1.0
dotnet 1.0.1 005db40cd1
"System.GC.Server": true


Fri, 17 Mar 2017 18:23:29 GMT

MAKE:
cp knucleotide.fsharpcore Program.fs
cp Include/fsharpcore/tmp.fsproj .
cp Include/fsharpcore/runtimeconfig.template.json .
mkdir obj
cp Include/fsharpcore/project.assets.json ./obj
cp Include/fsharpcore/tmp.fsproj.nuget.g.props ./obj
cp Include/fsharpcore/tmp.fsproj.nuget.g.targets ./obj
/usr/bin/dotnet build -c Release
Microsoft (R) Build Engine version 15.1.548.43366
Copyright (C) Microsoft Corporation. All rights reserved.

  tmp -> /home/dunham/benchmarksgame_quadcore/knucleotide/tmp/bin/Release/netcoreapp1.1/tmp.dll

Build succeeded.
    0 Warning(s)
    0 Error(s)

Time Elapsed 00:00:11.10
11.58s to complete and log all make actions

COMMAND LINE:
/usr/bin/dotnet ./bin/Release/netcoreapp1.1/tmp.dll 0 < knucleotide-input25000000.txt

PROGRAM OUTPUT:
A 30.295
T 30.151
C 19.800
G 19.754

AA 9.177
TA 9.132
AT 9.131
TT 9.091
CA 6.002
AC 6.001
AG 5.987
GA 5.984
CT 5.971
TC 5.971
GT 5.957
TG 5.956
CC 3.917
GC 3.911
CG 3.909
GG 3.902

1471758	GGT
446535	GGTA
47336	GGTATT
893	GGTATTTTAATT
893	GGTATTTTAATTTATAGT