diff --git a/mmtf.cabal b/mmtf.cabal --- a/mmtf.cabal +++ b/mmtf.cabal @@ -1,5 +1,5 @@ name: mmtf -version: 0.1.1.0 +version: 0.1.2.0 synopsis: Macromolecular Transmission Format implementation description: Haskell implementation of MMTF biological structure format. homepage: https://github.com/zmactep/mmtf#readme @@ -34,12 +34,24 @@ test-suite mmtf-test type: exitcode-stdio-1.0 hs-source-dirs: test + , src main-is: Spec.hs + other-modules: Bio.MMTF + , Bio.MMTF.Type + , Bio.MMTF.MessagePack + , Bio.MMTF.Decode + , Bio.MMTF.Decode.Codec + , Bio.MMTF.Decode.MessagePack build-depends: base , mmtf - , QuickCheck >= 2.9.2 && < 2.10 - , hspec >= 2.4.1 && < 2.5 + , QuickCheck >= 2.9.2 && < 2.13 + , hspec >= 2.4.1 && < 2.7 + , data-msgpack + , text , bytestring + , binary + , containers + , http-conduit ghc-options: -threaded -rtsopts -with-rtsopts=-N default-language: Haskell2010 diff --git a/src/Bio/MMTF/Decode.hs b/src/Bio/MMTF/Decode.hs --- a/src/Bio/MMTF/Decode.hs +++ b/src/Bio/MMTF/Decode.hs @@ -7,6 +7,7 @@ import Bio.MMTF.Type import Control.Monad ((>=>)) +import Data.ByteString.Lazy (empty) import Data.Map.Strict (Map) import Data.MessagePack (Object) import Data.Text (Text) @@ -27,76 +28,76 @@ -- chainData :: Monad m => Map Text Object -> m ChainData chainData mp = do gpc <- atP mp "groupsPerChain" asIntList - cil <- codec5 . parseBinary <$> atP mp "chainIdList" asBinary - cnl <- (codec5 . parseBinary <$>) <$> atPM mp "chainNameList" asBinary + cil <- codec5 . parseBinary <$> atP mp "chainIdList" asBinary + cnl <- codec5 . parseBinary <$> atPMD mp "chainNameList" asBinary empty pure $ ChainData gpc cil cnl -- | Parses atom data from ObjectMap -- atomData :: Monad m => Map Text Object -> m AtomData -atomData mp = do ail' <- (codec8 . parseBinary <$>) <$> atPM mp "atomIdList" asBinary - all' <- (codec6 . parseBinary <$>) <$> atPM mp "altLocList" asBinary - bfl' <- (codec10 . parseBinary <$>) <$> atPM mp "bFactorList" asBinary - xcl' <- codec10 . parseBinary <$> atP mp "xCoordList" asBinary - ycl' <- codec10 . parseBinary <$> atP mp "yCoordList" asBinary - zcl' <- codec10 . parseBinary <$> atP mp "zCoordList" asBinary - ol' <- (codec9 . parseBinary <$>) <$> atPM mp "occupancyList" asBinary +atomData mp = do ail' <- codec8 . parseBinary <$> atPMD mp "atomIdList" asBinary empty + all' <- codec6 . parseBinary <$> atPMD mp "altLocList" asBinary empty + bfl' <- codec10 . parseBinary <$> atPMD mp "bFactorList" asBinary empty + xcl' <- codec10 . parseBinary <$> atP mp "xCoordList" asBinary + ycl' <- codec10 . parseBinary <$> atP mp "yCoordList" asBinary + zcl' <- codec10 . parseBinary <$> atP mp "zCoordList" asBinary + ol' <- codec9 . parseBinary <$> atPMD mp "occupancyList" asBinary empty pure $ AtomData ail' all' bfl' xcl' ycl' zcl' ol' -- | Parses group data from ObjectMap -- groupData :: Monad m => Map Text Object -> m GroupData groupData mp = do gl' <- atP mp "groupList" asObjectList >>= traverse (transformObjectMap >=> groupType) - gtl' <- codec4 . parseBinary <$> atP mp "groupTypeList" asBinary - gil' <- codec8 . parseBinary <$> atP mp "groupIdList" asBinary - ssl' <- (map ssDec . codec2 . parseBinary <$>) <$> atPM mp "secStructList" asBinary - icl' <- (codec6 . parseBinary <$>) <$> atPM mp "insCodeList" asBinary - sil' <- (codec8 . parseBinary <$>) <$> atPM mp "sequenceIndexList" asBinary + gtl' <- codec4 . parseBinary <$> atP mp "groupTypeList" asBinary + gil' <- codec8 . parseBinary <$> atP mp "groupIdList" asBinary + ssl' <- fmap ssDec . codec2 . parseBinary <$> atPMD mp "secStructList" asBinary empty + icl' <- codec6 . parseBinary <$> atPMD mp "insCodeList" asBinary empty + sil' <- codec8 . parseBinary <$> atPMD mp "sequenceIndexList" asBinary empty pure $ GroupData gl' gtl' gil' ssl' icl' sil' -- | Parses group type from ObjectMap -- groupType :: Monad m => Map Text Object -> m GroupType groupType mp = do fcl' <- atP mp "formalChargeList" asIntList - anl' <- atP mp "atomNameList" asStrList - el' <- atP mp "elementList" asStrList - bal' <- atP mp "bondAtomList" asIntList - bol' <- atP mp "bondOrderList" asIntList - gn' <- atP mp "groupName" asStr + anl' <- atP mp "atomNameList" asStrList + el' <- atP mp "elementList" asStrList + bal' <- atP mp "bondAtomList" asIntList + bol' <- atP mp "bondOrderList" asIntList + gn' <- atP mp "groupName" asStr slc' <- atP mp "singleLetterCode" asChar - cct' <- atP mp "chemCompType" asStr + cct' <- atP mp "chemCompType" asStr pure $ GroupType fcl' anl' el' bal' bol' gn' slc' cct' -- | Parses structure data from ObjectMap -- structureData :: Monad m => Map Text Object -> m StructureData -structureData mp = do ttl' <- atPM mp "title" asStr - sid' <- atPM mp "structureId" asStr - dd' <- atPM mp "depositionDate" asStr - rd' <- atPM mp "releaseDate" asStr - nb' <- atP mp "numBonds" asInt - na' <- atP mp "numAtoms" asInt - ng' <- atP mp "numGroups" asInt - nc' <- atP mp "numChains" asInt - nm' <- atP mp "numModels" asInt - sg' <- atPM mp "spaceGroup" asStr - uc' <- (>>= ucDec) <$> atPM mp "unitCell" asFloatList - nol' <- ((>>= asFloatList) <$>) <$> atPM mp "ncsOperatorList" asObjectList - bal' <- (>>= traverse (transformObjectMap >=> bioAssembly)) <$> atPM mp "bioAssemblyList" asObjectList - el' <- (>>= traverse (transformObjectMap >=> entity)) <$> atPM mp "entityList" asObjectList - res' <- atPM mp "resolution" asFloat - rf' <- atPM mp "rFree" asFloat - rw' <- atPM mp "rWork" asFloat - em' <- atPM mp "experimentalMethods" asStrList - btl' <- (codec4 . parseBinary <$>) <$> atPM mp "bondAtomList" asBinary - bol' <- (codec2 . parseBinary <$>) <$> atPM mp "bondOrderList" asBinary +structureData mp = do ttl' <- atPMD mp "title" asStr "" + sid' <- atPMD mp "structureId" asStr "" + dd' <- atPMD mp "depositionDate" asStr "" + rd' <- atPMD mp "releaseDate" asStr "" + nb' <- atP mp "numBonds" asInt + na' <- atP mp "numAtoms" asInt + ng' <- atP mp "numGroups" asInt + nc' <- atP mp "numChains" asInt + nm' <- atP mp "numModels" asInt + sg' <- atPMD mp "spaceGroup" asStr "" + uc' <- (>>= ucDec) <$> atPM mp "unitCell" asFloatList + nol' <- (>>= asFloatList) <$> atPMD mp "ncsOperatorList" asObjectList [] + bal' <- atPMD mp "bioAssemblyList" asObjectList [] >>= traverse (transformObjectMap >=> bioAssembly) + el' <- atPMD mp "entityList" asObjectList [] >>= traverse (transformObjectMap >=> entity) + res' <- atPM mp "resolution" asFloat + rf' <- atPM mp "rFree" asFloat + rw' <- atPM mp "rWork" asFloat + em' <- atPMD mp "experimentalMethods" asStrList [] + btl' <- codec4 . parseBinary <$> atPMD mp "bondAtomList" asBinary empty + bol' <- codec2 . parseBinary <$> atPMD mp "bondOrderList" asBinary empty pure $ StructureData ttl' sid' dd' rd' nb' na' ng' nc' nm' sg' uc' nol' bal' el' res' rf' rw' em' btl' bol' -- | Parses bio assembly data from ObjectMap -- bioAssembly :: Monad m => Map Text Object -> m Assembly -bioAssembly mp = do nme' <- atP mp "name" asStr +bioAssembly mp = do nme' <- atP mp "name" asStr tlt' <- atP mp "transformList" asObjectList >>= traverse (transformObjectMap >=> transform) pure $ Assembly tlt' nme' @@ -104,14 +105,14 @@ -- transform :: Monad m => Map Text Object -> m Transform transform mp = do cil' <- atP mp "chainIndexList" asIntList - mtx' <- atP mp "matrix" asFloatList + mtx' <- atP mp "matrix" asFloatList pure $ Transform cil' mtx' -- | Parses entity data from ObjectMap -- entity :: Monad m => Map Text Object -> m Entity entity mp = do cil' <- atP mp "chainIndexList" asIntList - dsc' <- atP mp "description" asStr - tpe' <- atP mp "type" asStr - sqc' <- atP mp "sequence" asStr + dsc' <- atP mp "description" asStr + tpe' <- atP mp "type" asStr + sqc' <- atP mp "sequence" asStr pure $ Entity cil' dsc' tpe' sqc' diff --git a/src/Bio/MMTF/Decode/MessagePack.hs b/src/Bio/MMTF/Decode/MessagePack.hs --- a/src/Bio/MMTF/Decode/MessagePack.hs +++ b/src/Bio/MMTF/Decode/MessagePack.hs @@ -24,6 +24,12 @@ atPM :: Monad m => Map Text Object -> Text -> (Object -> m a) -> m (Maybe a) atPM m k conv = traverse conv $ M.lookup k m +atPMD :: Monad m => Map Text Object -> Text -> (Object -> m a) -> a -> m a +atPMD m k conv def = do x <- atPM m k conv + case x of + Just r -> pure r + Nothing -> pure def + asStr :: Monad m => Object -> m Text asStr (ObjectStr s) = pure s asStr _ = fail "Not a string data" diff --git a/src/Bio/MMTF/Type.hs b/src/Bio/MMTF/Type.hs --- a/src/Bio/MMTF/Type.hs +++ b/src/Bio/MMTF/Type.hs @@ -72,26 +72,26 @@ -- | Structure data -- -data StructureData = StructureData { title :: !(Maybe Text) -- ^ A short description of the structural data included in the file - , structureId :: !(Maybe Text) -- ^ An ID for the structure, for example the PDB ID if applicable - , depositionDate :: !(Maybe Text) -- ^ A date that relates to the deposition of the structure in a database - , releaseDate :: !(Maybe Text) -- ^ A date that relates to the release of the structure in a database +data StructureData = StructureData { title :: !Text -- ^ A short description of the structural data included in the file + , structureId :: !Text -- ^ An ID for the structure, for example the PDB ID if applicable + , depositionDate :: !Text -- ^ A date that relates to the deposition of the structure in a database + , releaseDate :: !Text -- ^ A date that relates to the release of the structure in a database , numBonds :: !Int32 -- ^ The overall number of bonds , numAtoms :: !Int32 -- ^ The overall number of atoms in the structure , numGroups :: !Int32 -- ^ The overall number of groups in the structure , numChains :: !Int32 -- ^ The overall number of chains in the structure , numModels :: !Int32 -- ^ The overall number of models in the structure - , spaceGroup :: !(Maybe Text) -- ^ The Hermann-Mauguin space-group symbol + , spaceGroup :: !Text -- ^ The Hermann-Mauguin space-group symbol , unitCell :: !(Maybe UnitCell) -- ^ Array of six values defining the unit cell - , ncsOperatorList :: !(Maybe [[Float]]) -- ^ List of lists representing 4x4 transformation matrices that are stored linearly in row major order (transformation matrices describe noncrystallographic symmetry operations needed to create all molecules in the unit cell) - , bioAssemblyList :: !(Maybe [Assembly]) -- ^ List of instructions on how to transform coordinates for an array of chains to create (biological) assemblies - , entityList :: !(Maybe [Entity]) -- ^ List of unique molecular entities within the structure + , ncsOperatorList :: ![[Float]] -- ^ List of lists representing 4x4 transformation matrices that are stored linearly in row major order (transformation matrices describe noncrystallographic symmetry operations needed to create all molecules in the unit cell) + , bioAssemblyList :: ![Assembly] -- ^ List of instructions on how to transform coordinates for an array of chains to create (biological) assemblies + , entityList :: ![Entity] -- ^ List of unique molecular entities within the structure , resolution :: !(Maybe Float) -- ^ The experimental resolution in Angstrom , rFree :: !(Maybe Float) -- ^ The R-free value , rWork :: !(Maybe Float) -- ^ The R-work value - , experimentalMethods :: !(Maybe [Text]) -- ^ List of experimental methods employed for structure determination - , bondAtomList :: !(Maybe [Int32]) -- ^ Pairs of values represent indices of covalently bonded atoms [binary (type 4)] - , bondOrderList :: !(Maybe [Int8]) -- ^ List of bond orders for bonds in 'bondAtomList' [binary (type 2)] + , experimentalMethods :: ![Text] -- ^ List of experimental methods employed for structure determination + , bondAtomList :: ![Int32] -- ^ Pairs of values represent indices of covalently bonded atoms [binary (type 4)] + , bondOrderList :: ![Int8] -- ^ List of bond orders for bonds in 'bondAtomList' [binary (type 2)] } deriving (Show, Eq) @@ -105,7 +105,7 @@ -- data ChainData = ChainData { groupsPerChain :: ![Int32] -- ^ List of the number of groups (aka residues) in each chain , chainIdList :: ![Text] -- ^ List of chain IDs [binary (type 5)] - , chainNameList :: !(Maybe [Text]) -- ^ List of chain names [binary (type 5)] + , chainNameList :: ![Text] -- ^ List of chain names [binary (type 5)] } deriving (Show, Eq) @@ -114,21 +114,21 @@ data GroupData = GroupData { groupList :: ![GroupType] -- ^ List of groupType objects , groupTypeList :: ![Int32] -- ^ List of pointers to 'groupType' entries in 'groupList' by their keys [binary (type 4)] , groupIdList :: ![Int32] -- ^ List of group (residue) numbers [binary (type 8)] - , secStructList :: !(Maybe [SecondaryStructure]) -- ^ List of secondary structure assignments [binary (type 2)] - , insCodeList :: !(Maybe [Char]) -- ^ List of insertion codes, one for each group (residue) [binary (type 6)] - , sequenceIndexList :: !(Maybe [Int32]) -- ^ List of indices that point into the sequence property of an entity object in the 'entityList' field that is associated with the chain the group belongs to [binary (type 8)] + , secStructList :: ![SecondaryStructure] -- ^ List of secondary structure assignments [binary (type 2)] + , insCodeList :: ![Char] -- ^ List of insertion codes, one for each group (residue) [binary (type 6)] + , sequenceIndexList :: ![Int32] -- ^ List of indices that point into the sequence property of an entity object in the 'entityList' field that is associated with the chain the group belongs to [binary (type 8)] } deriving (Show, Eq) -- | Atoms data -- -data AtomData = AtomData { atomIdList :: !(Maybe [Int32]) -- ^ List of atom serial numbers [binary (type 8)] - , altLocList :: !(Maybe [Char]) -- ^ List of alternate location labels, one for each atom [binary (type 6)] - , bFactorList :: !(Maybe [Float]) -- ^ List of atom B-factors in in A^2, one for each atom [binary (type 10)] +data AtomData = AtomData { atomIdList :: ![Int32] -- ^ List of atom serial numbers [binary (type 8)] + , altLocList :: ![Char] -- ^ List of alternate location labels, one for each atom [binary (type 6)] + , bFactorList :: ![Float] -- ^ List of atom B-factors in in A^2, one for each atom [binary (type 10)] , xCoordList :: ![Float] -- ^ List of x atom coordinates in A, one for each atom [binary (type 10)] , yCoordList :: ![Float] -- ^ List of y atom coordinates in A, one for each atom [binary (type 10)] , zCoordList :: ![Float] -- ^ List of z atom coordinates in A, one for each atom [binary (type 10)] - , occupancyList :: !(Maybe [Float]) -- ^ List of atom occupancies, one for each atom [binary (type 9)] + , occupancyList :: ![Float] -- ^ List of atom occupancies, one for each atom [binary (type 9)] } deriving (Show, Eq) diff --git a/test/Spec.hs b/test/Spec.hs --- a/test/Spec.hs +++ b/test/Spec.hs @@ -2,13 +2,40 @@ import Bio.MMTF import qualified Data.ByteString.Lazy as B +import Data.Int (Int8) import Test.Hspec +import Bio.MMTF.Decode.Codec + +codecSpec :: Spec +codecSpec = + describe "MMTF decoding" $ do + it "unpacks by Run-length encoding" $ do + let sample = [ 1, 10, 2, 1, 1, 4 ] :: [Int8] + runLengthDec sample `shouldBe` ([ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1 ] :: [Int8]) + it "unpacks by Delta encoding" $ do + let sample = [ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1 ] :: [Int8] + deltaDec sample `shouldBe` ([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16 ] :: [Int8]) + it "unpacks by Recursive indexing encoding" $ do + let sample = [ 127, 41, 34, 1, 0, -50, -128, 0, 7, 127, 0, 127, 127, 14 ] :: [Int8] + recIndexDec sample `shouldBe` [ 168, 34, 1, 0, -50, -128, 7, 127, 268 ] + it "upacks by Integer encoding" $ do + let sample = [ 100, 100, 100, 100, 50, 50 ] :: [Int8] + integerDec 100 sample `shouldBe` [ 1.00, 1.00, 1.00, 1.00, 0.50, 0.50 ] + +parserSpec :: Spec +parserSpec = + describe "MMTF parser" $ + it "should parse 1FSD" $ do + m <- fetch "1FSD" + (structureId . structure) m `shouldBe` "1FSD" + (numModels . structure) m `shouldBe` 41 + (length . bFactorList . atom) m `shouldBe` 20664 + (experimentalMethods . structure) m `shouldBe` ["SOLUTION NMR"] + (head . xCoordList . atom) m `shouldBe` (-12.847) + (last . xCoordList . atom) m `shouldBe` 5.672 + main :: IO () -main = hspec $ - describe "MMTF" $ - it "should parse 1FSD" $ do - contents <- B.readFile "resource/1FSD.mmtf" - m <- decode contents - (structureId . structure) m `shouldBe` Just "1FSD" - (numModels . structure) m `shouldBe` 41 +main = hspec $ do + codecSpec + parserSpec