Bioinformatics.Protein.hs

module Bioinformatics.Protein (
  AminoAcid (..),
  Protein,
  charToAminoAcid,
  rnaCodonTable,
  rnaToAminoAcids,
  rnaToProtein,
  monoisotopicMasses,
  openReadingFrame
) where

import Data.Char
import qualified Data.Map as M
import qualified Data.Maybe as Mb
import qualified Bioinformatics.RNANucleotide as R

data AminoAcid = A | C | D | E | F | G | H | I | K | L | M | N | P | Q | R | S | T | V | W | Y | Stop
                 deriving (Show, Eq, Ord, Read)

type Protein = [AminoAcid]

charToAminoAcid :: Char -> AminoAcid
charToAminoAcid c = read [toUpper c] :: AminoAcid


codonMapping = [([R.U,R.U,R.U], F),([R.C,R.U,R.U], L),([R.A,R.U,R.U], I),([R.G,R.U,R.U], V),
                ([R.U,R.U,R.C], F),([R.C,R.U,R.C], L),([R.A,R.U,R.C], I),([R.G,R.U,R.C], V),
                ([R.U,R.U,R.A], L),([R.C,R.U,R.A], L),([R.A,R.U,R.A], I),([R.G,R.U,R.A], V),
                ([R.U,R.U,R.G], L),([R.C,R.U,R.G], L),([R.A,R.U,R.G], M),([R.G,R.U,R.G], V),
                ([R.U,R.C,R.U], S),([R.C,R.C,R.U], P),([R.A,R.C,R.U], T),([R.G,R.C,R.U], A),
                ([R.U,R.C,R.C], S),([R.C,R.C,R.C], P),([R.A,R.C,R.C], T),([R.G,R.C,R.C], A),
                ([R.U,R.C,R.A], S),([R.C,R.C,R.A], P),([R.A,R.C,R.A], T),([R.G,R.C,R.A], A),
                ([R.U,R.C,R.G], S),([R.C,R.C,R.G], P),([R.A,R.C,R.G], T),([R.G,R.C,R.G], A),
                ([R.U,R.A,R.U], Y),([R.C,R.A,R.U], H),([R.A,R.A,R.U], N),([R.G,R.A,R.U], D),
                ([R.U,R.A,R.C], Y),([R.C,R.A,R.C], H),([R.A,R.A,R.C], N),([R.G,R.A,R.C], D),
                ([R.U,R.A,R.A], Stop),([R.C,R.A,R.A], Q),([R.A,R.A,R.A], K),([R.G,R.A,R.A], E),
                ([R.U,R.A,R.G], Stop),([R.C,R.A,R.G], Q),([R.A,R.A,R.G], K),([R.G,R.A,R.G], E),
                ([R.U,R.G,R.U], C),([R.C,R.G,R.U], R),([R.A,R.G,R.U], S),([R.G,R.G,R.U], G),
                ([R.U,R.G,R.C], C),([R.C,R.G,R.C], R),([R.A,R.G,R.C], S),([R.G,R.G,R.C], G),
                ([R.U,R.G,R.A], Stop),([R.C,R.G,R.A], R),([R.A,R.G,R.A], R),([R.G,R.G,R.A], G),
                ([R.U,R.G,R.G], W),([R.C,R.G,R.G], R),([R.A,R.G,R.G], R),([R.G,R.G,R.G], G)]

rnaCodonTable = M.fromList codonMapping

monoisotopicMasses = M.fromList [(A, 71.03711), (C, 103.00919),(D, 115.02694),
                                  (E, 129.04259), (F, 147.06841), (G, 57.02146),
                                  (H, 137.05891), (I, 113.08406), (K, 128.09496),
                                  (L, 113.08406), (M, 131.04049), (N, 114.04293),
                                  (P, 97.05276), (Q, 128.05858), (R, 156.10111),
                                  (S, 87.03203), (T, 101.04768), (V, 99.06841),
                                  (W, 186.07931), (Y, 163.06333 )]

lookupProteins :: [[R.RNANucleotide]] -> [AminoAcid]
lookupProteins = foldr (\rs acc -> if Mb.isJust rs then (Mb.fromJust rs):acc else acc) [] . map (\c -> M.lookup c rnaCodonTable)

rnaToAminoAcids :: [R.RNANucleotide] -> [AminoAcid]
rnaToAminoAcids = lookupProteins . R.rnaStringToRNAChunks

rnaToProtein :: [R.RNANucleotide] -> Protein
rnaToProtein = filter (\s -> s /= Stop) . lookupProteins . R.rnaStringToRNAChunks

openReadingFrame :: Protein -> Maybe Protein
openReadingFrame p@(M:aas) =
  let b = break ((==) Stop) p
      fb = fst b
      sb = snd b
      endsInStop xs = if not (null xs) then head xs == Stop else False
  in  if endsInStop sb
        then Just fb
        else Nothing
openReadingFrame _ = Nothing

OpenReadingFrames.hs

import qualified Bioinformatics.DNANucleotide as D
import qualified Bioinformatics.RNANucleotide as R
import qualified Bioinformatics.Utilities as U
import qualified Bioinformatics.Protein as P
import qualified Bioinformatics.FASTA as F
import qualified Data.Map as M
import Data.Maybe
import qualified Data.List as DL
import qualified Data.Set as DS

getRNA :: String -> [R.RNANucleotide]
getRNA = map (U.dnaToRNA . D.charToDNANucleotide)

getComplementedRNA :: String -> [R.RNANucleotide]
getComplementedRNA = map (U.dnaToRNA . D.nucleotideComplement . D.charToDNANucleotide)

getProtein :: String -> P.Protein
getProtein = P.rnaToAminoAcids . getRNA

getComplementedProtein :: String -> P.Protein
getComplementedProtein = P.rnaToAminoAcids . getComplementedRNA

getReadingFrames :: String -> [P.Protein]
getReadingFrames cs =
  let ps = map getProtein $ DL.tails cs
      cps = map getComplementedProtein . DL.tails $ reverse cs
      pSet = DS.fromList . catMaybes $ map P.openReadingFrame (ps ++ cps)
  in DS.toList pSet