Error when trying to perform bootstrap (RAxML 8.1.3)

261 views
Skip to first unread message

Paul Larson

unread,
Sep 17, 2014, 12:33:45 PM9/17/14
to ra...@googlegroups.com
Our computing cluster recently upgraded to the newest version of RAxML and the following command which ran fine under the earlier version

raxml -f d -s sep17.phy -n sep17boots -m GTRGAMMAI -q parts.txt -T 5 -p 13241 -b 13241 -# 1000


Now give me the following error.

raxml: evaluateGenericSpecial.c:3433: evaluateIterative: Assertion `partitionLikelihood < 0.0' failed.
Aborted (core dumped)

Am I doing something wrong or is this a bug?


Thanks,
-Paul

Andre J. Aberer

unread,
Sep 17, 2014, 1:57:45 PM9/17/14
to ra...@googlegroups.com
Hi Paul,

this is somewhat a shot in the dark, but could you try the exact same
run with an additional
--no-bfgs
?

It may very well be related to the invar model or the bug may only occur
for certain bootstrap replicates...

--
Best regards,
Andre
--
Sent with my mu4e

Paul Larson

unread,
Sep 17, 2014, 2:19:06 PM9/17/14
to ra...@googlegroups.com
Thanks for the swift reply

I get the same message with --no-bfgs, and when I exclude Invariant site estimation from the model (with and without --no-bfgs)

-Paul

Andre J. Aberer

unread,
Sep 17, 2014, 2:59:13 PM9/17/14
to ra...@googlegroups.com
ok, then I fear, this is a case for Alexis. It may still take a while
until he gets to this.

Paul Larson

unread,
Sep 17, 2014, 3:03:34 PM9/17/14
to ra...@googlegroups.com
Ok, thanks again!
 
I will use another version for now.
-Paul 

Ingo Michalak

unread,
Sep 23, 2014, 10:07:03 AM9/23/14
to ra...@googlegroups.com
Hey together,
I want to add, that the same happens for my setup: Ubuntu laptop, AVX.PTHREADS version, normal ("-f a") analyses. Older version (8.0.26) works. Full error message:

Time for BS model parameter optimization 0.111993
raxml: evaluateGenericSpecial.c:3433: evaluateIterative: Assertion `partitionLikelihood < 0.0' failed.
Aborted (core dumped)

Best regards

Alexandros Stamatakis

unread,
Oct 3, 2014, 4:39:48 AM10/3/14
to ra...@googlegroups.com
dear paul and ingo,

could you send me the input datasets that generate this error please?

thanks,

alexis
--
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

Christian Wurzb

unread,
Nov 13, 2014, 8:22:10 AM11/13/14
to ra...@googlegroups.com
I am afraid I need to join here.

I got the same error with or without AVX with:

raxmlHPC-PTHREADS -s DNA.fasta -n out.tree -x 1234567 -p 1234567 -# 1000 -d -f a -m GTRGAMMA -T 2 -u

Time for BS model parameter optimization 11.140697
raxmlHPC-PTHREADS: evaluateGenericSpecial.c:3433: evaluateIterative: Assertion `partitionLikelihood < 0.0' failed.
Aborted (core dumped)

Strangely, when I use -m GTRCAT this error does not occur.

Thanks,
Christian

Alexandros Stamatakis

unread,
Nov 13, 2014, 9:00:33 AM11/13/14
to ra...@googlegroups.com
Dear All,

Thanks for reporting, I have a good idea where the error might come
from, but I am completely overloaded right now, i.e., no time to write
code, just use release v8.0.26 that will work (hopefully).

I'll let you know once this is fixed.

Alexis

Alexandros Stamatakis

unread,
Nov 17, 2014, 5:14:44 AM11/17/14
to ra...@googlegroups.com
Dear Paul,

I just released version 8.1.4 of RAxML at
https://github.com/stamatak/standard-RAxML/releases this should fix the bug.

I'd appreciate if you could double-check.

Thanks,

Alexis

Alexandros Stamatakis

unread,
Nov 17, 2014, 5:17:11 AM11/17/14
to ra...@googlegroups.com
Dear Ingo,

I just released version 8.1.4 of RAxML at
https://github.com/stamatak/standard-RAxML/releases this should fix the bug.

I'd appreciate if you could double-check.

Thanks,

Alexis

Alexandros Stamatakis

unread,
Nov 17, 2014, 5:17:30 AM11/17/14
to ra...@googlegroups.com
Dear Christian,

I just released version 8.1.4 of RAxML at
https://github.com/stamatak/standard-RAxML/releases this should fix the bug.

I'd appreciate if you could double-check.

Thanks,

Alexis

On 13.11.2014 14:22, Christian Wurzb wrote:

Christian Wurzb

unread,
Nov 28, 2014, 3:40:35 AM11/28/14
to ra...@googlegroups.com
Thanks a lot Alexis, Problem solved!

Alexandros Stamatakis

unread,
Nov 28, 2014, 10:10:03 AM11/28/14
to ra...@googlegroups.com
thanks :-)

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

kin onn chan

unread,
Apr 22, 2015, 4:21:13 PM4/22/15
to ra...@googlegroups.com
Dear Alexis,

I am using your latest release 8.1.18 and got the same error:

evaluateGenericSpecial.c:3433: evaluateIterative: Assertion `partitionLikelihood < 0.0' failed

Should I revert back to 8.1.4? Many thanks in advance

Alexandros Stamatakis

unread,
Apr 23, 2015, 11:20:03 AM4/23/15
to ra...@googlegroups.com
Dear Kin,

Could you try the latest release v 8.1.20 first.

If this does not work either, please send me the input data and command
line to my personal email and I will have a look,

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>.
Reply all
Reply to author
Forward
0 new messages