Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

My end goal is a C++ implementation of

  • a Profile HMM (a profile hidden markov model is a model which contains information about a multiple sequence alignment of proteins; just providing some background as the type of model is not necessarily important for what I want to ask).
  • Baum Welch training (and implicitly Forward algorithm, Backward algorithm)

I have everything up and running as long as I don't try to use any silent states.

My biggest problem is that I do not know how to account/modify equations for silent states in my Forward/Backward algorithms. I didn't really find any good examples of this particular case online and I was wondering if anybody could give me a hint or point me in the right direction as I am not very math/stat savvy.

Follow up question: is there a user friendly package/toolbox with these algorithms already implemented? I ask this so that after I manage to complete my algos I can compare results to make sure I compute everything correctly.

I apologize if I was a bit incoherent about my problem. I will be monitoring this question and I will provide additional details as required.

share|improve this question
    
I saw this PDF today, it has some slides talking about silent states and how to modufy them in your code.. disi.unitn.it/~passerini/teaching/complex_systems/slides/… If you can do it, please, tell me. Thank you! –  user20497 Feb 5 '13 at 18:25
    
Could you please provide more details? –  Sven Hohenstein Feb 5 '13 at 18:53
    
e.g. hmmer.janelia.org This is the one I have experience with and it is used for bioinformatical research. There are many, many HMM implementations out there (C#, Java, python, C++, ...). Did you already search google for it and did not find something that worked for you (please specify why) or did you just not look yourself yet ;) -> sorry I just saw this post is very old, I don't know why it showed up at the top! –  ExpectoPatronum Jan 3 at 10:15
    
url is not working for me. –  ExpectoPatronum Jan 3 at 10:15

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.