Hello. /From my old course on HMM, I remember the theory being very simple (albeit with some not so simple maths). If this is the case, why are HMM programs apparently complex? Shouldn't one HMM program be just like another (being based on the same underlying principal)? It is just a guess, but I would have thought it would be possible to implement matching a sequence to a model in under 100 lines of perl? (I realize this is *minimal* functionality). Am I grossly overestimating my programming skills? Cheers, Dan.