Estimation of hundreds of gene trees for species tree analyses

[Jéssica Gillung]

Hi,

I have a very trivial question about gene tree estimation that is not specifically related to raxml or its usage; although raxml will ultimately be used. 

I have hundreds of loci as separate partitions, and I want to estimate a phylogenetic tree for each one of them. These gene trees will then be used as input in species tree estimation programs, such as Astral. I know that Raxml can only estimate one tree per alignment, but it will be a tedious task to run raxml on each gene partition separately (I have ~500 genes). So I thought that a loop script might do the trick, but I could not figure out how to write such script. Has any of you done that? Do you guys have any ideas on how I could have raxml automatically estimate a tree for each .phy alignment I have? 

This task has been done by multiple people over the years but I didn't find any help on the web. Indeed most species tree estimation publications do not explain the methods they used to generate the large numbers of gene trees used as input for the species tree estimation programs.

Thank you very much!

Jessica

[Alexandros Stamatakis]

hi jessica,

> I have hundreds of loci as separate partitions, and I want to estimate a
> phylogenetic tree for each one of them. These gene trees will then be
> used as input in species tree estimation programs, such as Astral. I
> know that Raxml can only estimate one tree per alignment,

that's not correct, it can conduct several tree searches per alignment,
check the -N option and the manual at:

http://sco.h-its.org/exelixis/resource/download/NewManual.pdf

> but it will be
> a tedious task to run raxml on each gene partition separately (I have
> ~500 genes). So I thought that a loop script might do the trick, but I
> could not figure out how to write such script. Has any of you done that?
> Do you guys have any ideas on how I could have raxml automatically
> estimate a tree for each .phy alignment I have?

Yes, that should be relatively easy, but you will have to do some
scripting, the way I would do it, assuming you are using a cluster:

write a perl or python script that will automatically generate job
submission scripts for all alignments ... check that the queue
submission files work correctly and then extend that script to also
automatically submit the job description files to the queuing system ...

hope that helps a bit,

alexis

>
> This task has been done by multiple people over the years but I didn't
> find any help on the web. Indeed most species tree estimation
> publications do not explain the methods they used to generate the large
> numbers of gene trees used as input for the species tree estimation
> programs.
>
> Thank you very much!
> Jessica
>

> --
> 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
Adjunct Professor, Dept. of Ecology and Evolutionary Biology, University
of Arizona at Tucson

www.exelixis-lab.org

[Jéssica Gillung]

Thank you for your help, Alexis!

[Carla Hurt]

Hi Jessica,

  I am trying to do the same thing.  I am not familiar with writing perl or python scripts (although I should be). Did you come up with a good way to do this?

Thank you,

Carla

[Alexandros Stamatakis]

Hi Carla,

Check the 3rd part wrapper scripts on:

https://sco.h-its.org/exelixis/web/software/raxml/index.html

especially:

raxml_wrapper.pl: A Perl script that reads a raxml.config file with
common run parameters and executes a directory of Phylip alignment files
in batch, then outputs the results in another directory. See the
documentation with "perldoc ./raxml_wrapper.pl".

You can extract the per-partition alignments from the supermatrix
directly with the respective RAxML command.

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