Speed Up Ancestral Genome Reconstruction?

[George Pacheco]

Dear RAxML Community,

I've been trying to reconstruct an ancestral genome with RAxML using the command:

raxmlHPC-PTHREADS -T 28 -f A -m GTRGAMMAI -s MEDIUM_40.chr17.fas -n MEDIUM_40.chr17.anc -t MEDIUM_40.B10k.nwk

However, it takes quite a bit of time (some estimates say around 6 months!) and I was wondering if there is anything I can do to speed it up.

Since I have several chromosomes I decided to do the reconstruction per chromosome.

Another issue is that there are quite a bit of parameters being estimated, so I thought of somehow reducing them:

- tried using a simpler model (--HKY85), but it didn't help much

- could also choose an average chromosome to estimate all the parameters, and then use these as fixed parameters to infer the ancestral sequences on the other chromosomes, but I don't think RAxML can do this, right?

Is there anything I can do to speed it up?

Many thanks in advance!

[Alexandros Stamatakis]

Dear George,

How large is your dataset in terms of #sites and #taxa?

Ancestral reconstruction is not time consuming in principle, what is
taking a lot of time here is presumably the model parameter optimization
that is done prior to the ancestral state reconstruction.

Trying on a per-chromosome basis is what I would have suggested as well.

Alternatively you may just have to wait a bit until that feature is
implemented in RAxML-NG by Alexey I am afraid.

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

[George Pacheco]

Dear Alexis,

I do appreciate your thoughts and suggestion.
I will proceed trying on a per-chromosome basis. 

Your best, George.