SumTrees common ancestor node ages

[Nitish Narula]

Hi everyone,

I am using sumtrees to summarize posterior sets from a few BEAST runs. In treeannotator application, which comes with BEAST, there is an option to set common ancestors (ca) heights. Is there is something similar in sumtrees? I can't seem to find it.

Thanks,

Nitish

[Jeet Sukumaran]

Do you mean to set the ages of the nodes to the mean of their ages in
the MCMC sample?

If so, the ``-e`` or ``--set-edges`` option will do this if you specify:

-e mean-age
--set-edges mean-age

If you also add:

--summarize-node-ages

you will get other information (range, min/max, 95% HPD) is the metadata.

-- jeet

> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to dendropy-user...@googlegroups.com
> <mailto:dendropy-user...@googlegroups.com>.
> For more options, visit https://groups.google.com/d/optout.

--



--------------------------------------
Jeet Sukumaran
--------------------------------------
jeetsu...@gmail.com
--------------------------------------
Blog/Personal Pages:
http://jeetworks.org/
GitHub Repositories:
http://github.com/jeetsukumaran
Photographs (as stream):
http://www.flickr.com/photos/jeetsukumaran/
Photographs (by galleries):
http://www.flickr.com/photos/jeetsukumaran/sets/
--------------------------------------

[Nitish Narula]

Hi Jeet, 

I am not sure if I'm looking for the mean-age option in the edge lengths argument. 

In treeannotator there is an option for heights argument called 'ca' (common ancestors heights) which, from what I understand, avoids negative length edges without collapsing to polytomies. There isn't clear documentation on it, so the best I can give you is this and this (btw, your documentation is markedly better!). The other options in treeannotator - median and mean heights - have the possibility of negative edges. 

In sumtrees, there are mean-age and median-age options (among others), which I thought would be similar to the mean and median heights options in treeannotator. But there isn't something similar to the ca heights option, at least not explicitly. Is the mean-age option similar to the ca heights option? I have used "--collapse-negative-edges" but that returns polytomies and I am trying to avoid that. 

Let me know if I'm off on my understanding of these height/age/edge length options.

Thanks,

Nitish

On Thursday, September 10, 2015 at 5:05:02 PM UTC-6, Jeet Sukumaran wrote:

Do you mean to set the ages of the nodes to the mean of their ages  in
the MCMC sample?

If so, the ``-e`` or ``--set-edges`` option will do this if you specify:

        -e mean-age
        --set-edges mean-age

If you also add:

        --summarize-node-ages

you will get other information (range, min/max, 95% HPD) is the metadata.

-- jeet

On 9/10/15 1:51 PM, Nitish Narula wrote:
> Hi everyone,
>
> I am using sumtrees to summarize posterior sets from a few BEAST runs.
> In treeannotator application, which comes with BEAST, there is an option
> to set common ancestors (ca) heights. Is there is something similar in
> sumtrees? I can't seem to find it.
>
> Thanks,
> Nitish
>
> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to dendropy-user...@googlegroups.com

> <mailto:dendropy-users+unsub...@googlegroups.com>.

[Jeet Sukumaran]

Ahh, OK.

Well, I am afraid SumTrees does not do this at this time.

However, it should be easy to add. If you submit an issue request on the
GitHub site, we will try and work on it.

-- jeet

On 9/10/15 7:34 PM, Nitish Narula wrote:
> Hi Jeet,
>
> I am not sure if I'm looking for the mean-age option in the edge lengths
> argument.
>
> In treeannotator there is an option for heights argument called 'ca'
> (common ancestors heights) which, from what I understand, avoids
> negative length edges without collapsing to polytomies. There isn't
> clear documentation on it, so the best I can give you is this

> <http://www.biomedcentral.com/1471-2148/13/221#sec2> and this
> <https://github.com/CompEvol/beast2/blob/master/src/beast/app/treeannotator/TreeAnnotator.java#L383>

> > an email to dendropy-user...@googlegroups.com <javascript:>
> > <mailto:dendropy-user...@googlegroups.com <javascript:>>.

> > For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
> --
>
>
>
> --------------------------------------
> Jeet Sukumaran
> --------------------------------------

> jeetsu...@gmail.com <javascript:>

> --------------------------------------
> Blog/Personal Pages:
> http://jeetworks.org/
> GitHub Repositories:

> http://github.com/jeetsukumaran <http://github.com/jeetsukumaran>

> Photographs (as stream):
> http://www.flickr.com/photos/jeetsukumaran/
> <http://www.flickr.com/photos/jeetsukumaran/>
> Photographs (by galleries):
> http://www.flickr.com/photos/jeetsukumaran/sets/

> <http://www.flickr.com/photos/jeetsukumaran/sets/>
> --------------------------------------

>
> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to dendropy-user...@googlegroups.com

> <mailto:dendropy-user...@googlegroups.com>.