RY coding?

[Jan Strugnell]

Dear BEAST users,

I was wondering how/if BEAST can handle an RY coded dataset? Will it
give sensible results with regard to divergence time estimation?

Many thanks,

Jan

[Andrew Rambaut]

Hi Jan,

By default, BEAST treats Rs and Ys (and other ambiguity codes) as
gaps (i.e., no information).
This speeds things up. However if you want to use ambiguity codes
properly, then you can
add the following attribute in the XML:

<treeLikelihood id="treeLikelihood" useAmbiguities="true"> ....

You will also need to remove the operator on kappa (as there are no
transitions) and give
it an initial value of 0.5 (equal rates under the HKY model).

Andrew

___________________________________________________________________
Andrew Rambaut
Institute of Evolutionary Biology University of Edinburgh
Ashworth Laboratories Edinburgh EH9 3JT
EMAIL - a.ra...@ed.ac.uk TEL - +44 131 6508624

[Andrew Rambaut]

By default, BEAST treats ambiguity codes as missing data - i.e., the same as a gap or ?. By doing this it can make an optimization that provides close to a 2-fold speed up. This is probably worth it if ambiguities are rare. If you want to use the information in ambiguities then you need to edit the XML, find the element <treeLikelihood ...> and change useAmbigities="false" to "true".

I don't think I am answering your question here but the above is necessary before you can explore the effects of what you are trying to do.

Andrew

On 19 Oct 2012, at 12:05, Daniele Salvi <danieles...@gmail.com> wrote:

Dear Andrew,

 

How does the use of ambiguity codes affect the GTR models setting?

I have several nuclear alignments for which I would like to use the information of the ambiguity codes (Y, R, M, K, S, W).

I’ll remove the operator kappa and set its initial value of 0.5 for the HKY models;

do I have to change other priors/operators setting of the GTR models?

 

Thanks in advance,

 

Daniele

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To view this discussion on the web visit https://groups.google.com/d/msg/beast-users/-/RJP8UDMoqZ8J.
To post to this group, send email to beast...@googlegroups.com.
To unsubscribe from this group, send email to beast-users...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/beast-users?hl=en.

[Daniele Salvi]

Thanks for your reply!

I'm aware that it will speed down my run, but in fact I have quite a lot of ambiguity codes and thus I would like not to loose this information. At the same time I would like to specify my models as correctly as possible.

So, my question was:

 

when setting the useAmbiguities to "true",  in practise should I modify also some GTR model parameters specification (priors and operators)?

 

And also, how this setting will affect the model parametersr estimation?

 

Thanks again for your help,

 

Daniele

--

Daniele Salvi, PhD

CIBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos

Campus Agrário de Vairão, Rua Padre Armando Quintas 4485-661, Vairão, Portugal

Mobile: +39-3331167735; +351-933530870

Personal Page on CIBIO's website

[Jacob Berv]

bump