Using "weighted" sequences for ML computations

[Ikram]

Hi everyone,

I was wondering if BPP supports ML computations, given a sequence (with or without a tree - let's say tree is given for the time being) where I weigh each site according to some previously computed score. In other words, the default weighting we use is uniform (each site has weight 1.0); consider using some lower or upper weights to model the "importance" of respective sites.

Is it currently possible in BPP?

Cheers,

Ikram

[Julien Yann Dutheil]

Hi Ikram

Actually the current ML implementation uses weights (the frequency of each site), so that it is easy to plug in your own weights. I am afraid no specific method has been developed for that purpose, but it should not be too much work to derive one ML class and add this functionality.

Best,

Julien.

--
You received this message because you are subscribed to the Google Groups "Bio++ Development Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biopp-devel-fo...@googlegroups.com.
To post to this group, send email to biopp-de...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/biopp-devel-forum/9cca800c-5e0d-4abb-9ab1-f8c608c37063%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Ikram]

You are right, apparently it shouldn't be much of work. Its just that before attempting, I was thinking not to reinvent the wheel... :)

Best Regards,

Ikram