diff --git a/ChangeLog.md b/ChangeLog.md --- a/ChangeLog.md +++ b/ChangeLog.md @@ -7,5 +7,10 @@ ## [Unreleased] +## 0.1.0.1 +* Small fixes in description fields read +* Documentation fixes +* Usage example in README + ## 0.1.0.0 * Any UniProt-KB file can be parsed diff --git a/README.md b/README.md --- a/README.md +++ b/README.md @@ -15,12 +15,24 @@ ## Usage example -To use the parser and types import: -``` -λ> import Bio.Uniprot +You can use this simple code to parse any UniProt file: + +``` haskell +import Bio.Uniprot +import Data.Attoparsec.Text (parseOnly) +import Data.Text (Text) +import qualified Data.Text.IO as TIO +import Text.Pretty.Simple (pPrint) + +main :: IO () +main = do + uniprot <- TIO.getContents + case parseOnly parseRecord uniprot of + Left err -> putStrLn "Error on parse" + Right obj -> pPrint obj ``` -Than you can parse any `Text` of UniProt by using `parseRecord` function. The result will be presented by +So you can parse any stdin `Text` of UniProt by using `parseRecord` function. The result will be presented by a `Record` datatype: ``` haskell data Record = Record @@ -42,4 +54,9 @@ , ft :: [FT] , sq :: SQ } deriving (Show, Eq, Ord) +``` + +You can run this to test the implementation: +``` bash +stack runhaskell uniprot.hs < example/LOLA2_DROME.dat ``` diff --git a/src/Bio/Uniprot/Parser.hs b/src/Bio/Uniprot/Parser.hs --- a/src/Bio/Uniprot/Parser.hs +++ b/src/Bio/Uniprot/Parser.hs @@ -3,7 +3,7 @@ {-# LANGUAGE TupleSections #-} module Bio.Uniprot.Parser where -import Prelude hiding (concat, init, null) +import Prelude hiding (init, null) import qualified Prelude as P (concat, id, init) import Bio.Uniprot.Type @@ -16,8 +16,6 @@ splitOn, unpack, unwords) import Data.Void -import Debug.Trace (trace) - -- |Describes possible name type of DE section. data NameType = RecName | AltName | SubName | Flags | None deriving (Show) @@ -212,6 +210,7 @@ char '.' pure OH{..} +-- |Parses RN, RP, RC, RX, RG, RA, RT and RL lines of UniProt-KB text file. parseRef :: Parser Reference parseRef = do rn <- parseRN @@ -225,7 +224,7 @@ rt <- optional (parseRT <* endOfLine) rl <- parseRL pure Reference{..} - where + where parseRN :: Parser Int parseRN = (string "RN [" *> decimal) <* char ']' @@ -260,6 +259,7 @@ string "RL " pack . P.init <$> parseMultiLineComment "RL" 3 +-- |Parses CC lines of UniProt-KB text file. parseCC :: Parser CC parseCC = do string "CC -!- " @@ -269,6 +269,7 @@ comment <- pack <$> parseMultiLineComment "CC" 7 pure CC{..} +-- |Parses DR lines of UniProt-KB text file. parseDR :: Parser DR parseDR = do string "DR " @@ -327,7 +328,7 @@ splitStr n acc ('(':xs) = splitStr (n+1) ('(':acc) xs splitStr n acc (')':xs) = splitStr (n-1) (')':acc) xs splitStr n acc (x:xs) = splitStr n (x:acc) xs - + -- |Parses SQ lines of UniProt-KB text file. parseSQ :: Parser SQ parseSQ = do @@ -352,28 +353,29 @@ -- |Parses whole UniProt-KB record. parseRecord :: Parser Record -parseRecord = Record <$> (parseID <* endOfLine) - <*> (parseAC <* endOfLine) - <*> (parseDT <* endOfLine) - <*> (parseDE <* endOfLine) - <*> (parseGN <* endOfLine) - <*> (parseOS <* endOfLine) - <*> optional (parseOG <* endOfLine) - <*> (parseOC <* endOfLine) - <*> optional (parseOX <* endOfLine) - <*> many' (parseOH <* endOfLine) - <*> many' (parseRef <* endOfLine) - <*> many' (parseCC <* endOfLine) - <*> many' (parseDR <* endOfLine) - <*> (parsePE <* endOfLine) - <*> (parseKW <* endOfLine) - <*> many' (parseFT <* endOfLine) - <*> (parseSQ <* endOfLine) - <* parseEnd +parseRecord = Record <$> (parseID <* endOfLine) + <*> (parseAC <* endOfLine) + <*> (parseDT <* endOfLine) + <*> (parseDE <* endOfLine) + <*> option [] (parseGN <* endOfLine) + <*> (parseOS <* endOfLine) + <*> optional (parseOG <* endOfLine) + <*> (parseOC <* endOfLine) + <*> (parseOX <* endOfLine) + <*> many' (parseOH <* endOfLine) + <*> many' (parseRef <* endOfLine) + <*> many' (parseCC <* endOfLine) + <*> many' (parseDR <* endOfLine) + <*> (parsePE <* endOfLine) + <*> optional (parseKW <* endOfLine) + <*> many' (parseFT <* endOfLine) + <*> (parseSQ <* endOfLine) + <* parseEnd -- = Helper parsers -- |Transforms any parser to a parser of maybe value. +-- -- >>> parseOnly (optional digit) "1" -- Right (Just 1) -- @@ -411,7 +413,7 @@ comm <- (:) <$> parseTillEnd <*> many' (do endOfLine string start - count (skip - 1) space -- leave one space to separate words + count (skip - 1) (char ' ') -- leave one space to separate words parseTillEnd) pure $ hyphenConcat comm @@ -429,6 +431,12 @@ hyphenConcat :: [String] -> String hyphenConcat [] = [] hyphenConcat [x] = x -hyphenConcat (x:y:ys) = if last x == '-' - then x ++ tail y ++ hyphenConcat ys - else x ++ y ++ hyphenConcat ys +hyphenConcat (x:y:ys) = x ++ hyphenConcat (sy:ys) + where + sy :: String + sy | last x == '-' = tail y + | isAA (last x) && isAA (y !! 1) = tail y + | otherwise = y + + isAA :: Char -> Bool + isAA = inClass "ACDEFGHIKLMNPQRSTVWY" diff --git a/src/Bio/Uniprot/Type.hs b/src/Bio/Uniprot/Type.hs --- a/src/Bio/Uniprot/Type.hs +++ b/src/Bio/Uniprot/Type.hs @@ -246,13 +246,13 @@ , os :: OS , og :: Maybe OG , oc :: OC - , ox :: Maybe OX + , ox :: OX , oh :: [OH] , refs :: [Reference] , cc :: [CC] , dr :: [DR] , pe :: PE - , kw :: KW + , kw :: Maybe KW , ft :: [FT] , sq :: SQ } deriving (Show, Eq, Ord) diff --git a/test/Spec.hs b/test/Spec.hs --- a/test/Spec.hs +++ b/test/Spec.hs @@ -941,6 +941,6 @@ it "parses PD1 example" $ do let record = Record idAns acAns dtAns deAns gnAns osAns Nothing ocAns - (Just oxAns) [] refAns ccAns - drAns peAns kwAns ftAns sqAns + oxAns [] refAns ccAns + drAns peAns (Just kwAns) ftAns sqAns parseOnly parseRecord pd1Str `shouldBe` Right record diff --git a/uniprot-kb.cabal b/uniprot-kb.cabal --- a/uniprot-kb.cabal +++ b/uniprot-kb.cabal @@ -2,10 +2,10 @@ -- -- see: https://github.com/sol/hpack -- --- hash: 35141853a683c97f0604c324c743203061f6bba440e50259a2f29dcc8d8825cc +-- hash: 46801dc834178bde36de5bfcd64e4b397b76da7ac4dbe6f22b7275f5a1661110 name: uniprot-kb -version: 0.1.0.0 +version: 0.1.0.1 synopsis: UniProt-KB format parser description: Specification implementation of https://web.expasy.org/docs/userman.html category: Bio