IC (Internode cofidence) / TCA from partial (taxa) gene trees

[Matthew Brown]

HI Kassian and Alexandros, 

I am trying to do some IC/TCA analyses with RAxML on a phylogenomic dataset that I developed. I am trying to do IC analyses on this dataset, but it seems that RAxML (v8.2.6) does not allow for incomplete taxon sampling in gene trees using the -L option. I am wondering if I am missing something, as it says partial is allowable from 8.2.0 on. In the manual on pg49 it states "Also note that, as of version 8.2.0 We can conduct TC/IC calculations on collections of partial gene trees. The corrections we apply for partial gene trees are described in this paper here: http://dx.doi.org/10.1101/022053". 

Thank you 
Matt

[Matthew Brown]

I figured out the problem, which was in my call of Raxml, so there is no need to reply.

needed the option "-f i"

Thank you

Matt

[Matthew Brown]

I still have a question if MR and MRE trees are still able to be constructed from partial input trees. If you run -L MRE it will core dump.

Thanks,

Matt

[Alexandros Stamatakis]

Dear Matt,

Thanks for answering your own question :-)

Unfortunately, building MR and MRE trees works only on sets of trees
where each tree in the set contains all taxa.

Alexis

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

--
Alexandros (Alexis) Stamatakis

Research Group Leader, Heidelberg Institute for Theoretical Studies
Full Professor, Dept. of Informatics, Karlsruhe Institute of Technology

www.exelixis-lab.org

[Matthew Brown]

Hi Alexis, 

One final question which I can not seem to find the answer to. 

When using -f i with a ML tree and a set of partial (taxa) gene trees, what do the branch labels represent. It is not clear if they are IC and TC or TC and IC. I am pretty sure they are IC and TC, but I would like to confirm this. 

Thanks, 

Matt

[Alexandros Stamatakis]

They are IC, TC is the score over the entire tree.

Alexis

> > an email to raxml+un...@googlegroups.com <javascript:>
> > <mailto:raxml+un...@googlegroups.com <javascript:>>.

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

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

>
> --
> Alexandros (Alexis) Stamatakis
>
> Research Group Leader, Heidelberg Institute for Theoretical Studies
> Full Professor, Dept. of Informatics, Karlsruhe Institute of Technology
>

> www.exelixis-lab.org <http://www.exelixis-lab.org>