-- Hoogle documentation, generated by Haddock
-- See Hoogle, http://www.haskell.org/hoogle/


-- | RNA folding training data
--   
--   Parameter training for RNA secondary structure prediction tools
--   requires data to train on. Since there are a number of different
--   available formats, and handling them all in the training tools is a
--   pain, we have this library and programs. <a>MkTrainingData</a>
--   transforms different formats and they all produce a common <a>training
--   data</a> format. This format is Haskell-readable (and only partially
--   human-readable) line-by-line. Generating additional training data is
--   therefor easy as one can just <a>cat</a> together different training
--   files.
--   
--   Note that several features are designed around <i>extended</i> RNA
--   secondary structures.
--   
--   Now with some filtering and manipulation options.
@package BiobaseTrainingData
@version 0.1.2.2

module Biobase.TrainingData

-- | One training data element. We can store sequence and known structure
--   (primary, secondary) as well as a predicted structure (stored, if
--   <a>not.null</a>). The <a>weight</a> is how strongly this element
--   should influence a training system. <a>extendedKnowledge</a> is True,
--   if the data element comes from a source which knows about extended
--   secondary structures, like the PDB. Otherwise it is False.
--   
--   NOTE During training, one should not penalize non-canonical
--   predictions in interior loops and multibranch loops, unless they
--   hinder formation of true pairs.
--   
--   TODO at some point we will move toward pseudoknots and other fun
data TrainingData
TrainingData :: String -> [ExtPairIdx] -> [ExtPairIdx] -> Double -> Bool -> [String] -> TrainingData
primary :: TrainingData -> String
secondary :: TrainingData -> [ExtPairIdx]
predicted :: TrainingData -> [ExtPairIdx]
weight :: TrainingData -> Double
extendedKnowledge :: TrainingData -> Bool
comments :: TrainingData -> [String]

-- | Create <a>TrainingData</a> from various sources.
class MkTrainingData a
mkTrainingData :: MkTrainingData a => a -> TrainingData
instance [overlap ok] Read TrainingData
instance [overlap ok] Show TrainingData
instance [overlap ok] MkTrainingData RNAstrand
instance [overlap ok] MkTrainingData LinFR3D

module Biobase.TrainingData.Manip

-- | Left elements are filtered out, Right elements are kept.
type TDmanip = Either TrainingData TrainingData


-- | Prospective <a>TrainingData</a> elements need to be filtered as there
--   are a number of entries which do not provide good training.
module Biobase.TrainingData.Filter

-- | Filter out elements containing not enough base pairs (in relative
--   terms)
fMinRelPairs :: Maybe Double -> TDmanip -> TDmanip

-- | Error-checking filter.
fErrorCheck :: TDmanip -> TDmanip


-- | Fancy importer for TrainingData.
module Biobase.TrainingData.Import

-- | Enumeratee producing <a>TrainingData</a> from a bytestring.
eneeTrainingData :: Monad m => Enumeratee ByteString [TrainingData] m a

-- | Convenience function
fromFile :: FilePath -> IO [TrainingData]