Strange Tree Toplogy

[Austin Varela]

Hello,

I am using RAxML v8.2.11 for generating branch lengths, however for one of the generated trees, it actually changes the topology--despite me passing a consensus tree topology to it. Is this intended?

More info:

I am calling it via:

raxmlHPC-PTHREADS -s my_concatenated_alignments.fasta -n name -p 12345 -f e -m PROTGAMMAAUTO -t my_consensus_tree.nwk -T 12 -w /my/working/dir

 But as you can see the generated output has incorrect placements such as these:

When compared to my consensus tree topology:


Thanks in advance!

[Alexandros Stamatakis]

That's weird, this command should not change the tree topology.

Could you send me all input files to my personal email address such that
I can check?

Please consider switching to RAxML-NG
https://github.com/amkozlov/raxml-ng as support for standard RAxML will
be continuously reduced.

Alexis

On 16.05.2018 17:51, Austin Varela wrote:
> Hello,
>
> I am using RAxML v8.2.11 for generating branch lengths, however for one
> of the generated trees, it actually changes the topology--despite me
> passing a consensus tree topology to it. Is this intended?
>
> More info:
>
> I am calling it via:
>
> raxmlHPC-PTHREADS -s my_concatenated_alignments.fasta -n name -p
> 12345 -f e -m PROTGAMMAAUTO -t my_consensus_tree.nwk -T 12 -w
> /my/working/dir
>
>
>  But as you can see the generated output has incorrect placements such
> as these:

> https://puu.sh/AngrE/91138b8238.png
> When compared to my consensus tree topology:

> https://puu.sh/AngsW/72f91ffb04.png
>
> Thanks in advance!
>

> --
> 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

[Austin Varela]

Alright, I have sent you the email.

[Alexandros Stamatakis]

Dear Austin,

Just checked: your input tree does not contain all taxa, thus the ones
that you showed on your screenshot will become resolved via maximum
likelihood without any constraints, hence the difference in tree
topologies you are observing.

I executed RAxML as follows:

./raxmlHPC-PTHREADS-AVX -T 2 -s mail/concatenated.fasta -n nMRP -f e -m
PROTGAMMAAUTO -t mail/pruned_tree.nwk

and got the following warning:

You provided an incomplete starting tree 82 alignmnet has 84 taxa
Error: you need to specify a random number seed with "-p" for the
randomized stepwise addition
parsimony algorithm or random tree building algorithm such that runs can
be reproduced and debugged ... exiting
raxmlHPC-PTHREADS-AVX: fastDNAparsimony.c:137: checkSeed: Assertion
`adef->parsimonySeed > 0' failed.

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>

[Austin Varela]

Ah you are correct. It looks like my consensus tree pre-processing dropped them for some reason. Thanks so much for your help!

[Austin Varela]

Just an update, I have removed the missing taxa from my alignment and I am still having the same issue.

[Alexandros Stamatakis]

This can't be, are you sure you removed the correct taxa?

alexis

[Austin Varela]

It turns out I actually pruned the wrong taxa, now everything is working perfectly. Sorry for the trouble!