Christian Hoener zu Siederdissen <choe...@tbi.univie.ac.at> writes: > Ketils basic typedefs should suit us well for sequence things. I will > have a type class MkPrimary with (mkPrimary :: SeqData -> VU.Vector > Nuc), for example.
Moreover, if you define an instance BioSeq for VU.Vector Nuc, you get to deal with them in a generic way - including outputting them as Fast[aq]. In retrospect, this is a great but obvious improvement over biolib, which dealt with sequences through the 'Sequence t' data type. > Btw. if you need performance, consider staying away from some types > like Word. There are fun open bugs, where Int is 2.5x faster than > Word. ;-) Currently, sequence qualities are Word8s - based on their range which is 0..something (where something isn't entirely clear, but certainly less than 256), and on fitting them into ByteStrings. Now, it may make sense to use an Int instead, for instance, overflows are a perpetual source of errors. That would involve a fromIntegral in every access, so I'm not sure it's a net win. Anyway, thanks for the input so far, I think this has been a move in the right direction, all that's left is to follow it up... -k -- If I haven't seen further, it is by standing in the footprints of giants _______________________________________________ Biohaskell mailing list Biohaskell@biohaskell.org http://malde.org/cgi-bin/mailman/listinfo/biohaskell
