[Biojava-l] Partial path probs

Schreiber, Mark mark.schreiber at agresearch.co.nz
Fri Jul 18 11:08:32 EDT 2003


> An alternate approach, which BioJava *will* help you with, is 
> to calculate both the forward and the backward DP matrices, 
> then multiply these together and normalize by the overall 
> backwards/forwards probability.  This then tells you the 

By normalize do you mean divide each value by the total forward or
backward probability? So by doing this I would expect that the
forward-backward prob should equal the forward and the backward probs?

Also, can someone enlighten me how the matrices are indexed (there is
not much javadoc in DP)? Currently I'm assuming that the x axis is the
array returned by getStates and the y axis is the aligned SymbolList
from that state path, with a magical state at each end but I'm not too
clear on this.

All these questions are leading up to the development of some semi
markov models or emissions with duration. Also should this be on
biojava-dev or are others finding this useful?

- Mark
=======================================================================
Attention: The information contained in this message and/or attachments
from AgResearch Limited is intended only for the persons or entities
to which it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipients is prohibited by AgResearch
Limited. If you have received this message in error, please notify the
sender immediately.
=======================================================================



More information about the Biojava-l mailing list