RAxML 8.2.7 crashes during final model optimization
846 views
Skip to first unread message
fsc...@berkeley.edu
Sep 9, 2016, 12:38:13 AM9/9/16
to raxml
Dear Alexis,
We are trying to run a large tree with 17492 nucleic acid sequences and 1533 alignment patterns using raxmlHPC-HYBRID-SSE3 (8.2.7) compiled (gcc version 4.8.5) on our cluster (each node has 256 Gb memory) with openmpi/1.8.7:
mpirun --np 20 --map-by node $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p 12345 -N 20 -m GTRCATI -n $suffix -C -T 16It runs without any issues on a smaller data set (~2000 sequences). However, with the large data set it crashes during final model optimization:
Conducting final model optimizations on all 20 trees under GAMMA-based models ....
#######
raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion `rate >= 0.0001 && rate <= 1000000.0' failed.
We also tested the "-F" option:
Base frequencies: 0.254 0.230 0.313 0.203
#######
Likelihood problem in model optimization l1: -inf l2: -8438342.1678280774503946304321289062500000000000 tolerance: 0.0000084383421678280775418690140554645041
--------------------------------------------------------------------------
mpirun noticed that process rank 5 with PID 0 on node mc0208-ib.nersc.gov exited on signal 6 (Aborted).
--------------------------------------------------------------------------
And on the Cipres portal:
raxmlHPC-HYBRID -T 4 -f a -N autoMRE -n result -s infile.txt -c 25 -p 12345 -m GTRCATI -x 12345raxmlHPC-HYBRID: rapidBootstrap.c:432: treeOptimizeThorough: Assertion `res > 0' failed.
[comet-07-15:16651] *** Process received signal ***
[comet-07-15:16651] Signal: Aborted (6)
[comet-07-15:16651] Signal code: (-6)
[comet-07-15:16651] [ 0] /lib64/libpthread.so.0[0x34d440f7e0]
[comet-07-15:16651] [ 1] /lib64/libc.so.6(gsignal+0x35)[0x34d3c32625]
[comet-07-15:16651] [ 2] /lib64/libc.so.6(abort+0x175)[0x34d3c33e05]
[comet-07-15:16651] [ 3] /lib64/libc.so.6[0x34d3c2b74e]
[comet-07-15:16651] [ 4] /lib64/libc.so.6(__assert_perror_fail+0x0)[0x34d3c2b810]
[comet-07-15:16651] [ 5] raxmlHPC-HYBRID[0x493b11]
[comet-07-15:16651] [ 6] raxmlHPC-HYBRID[0x44821c]
[comet-07-15:16651] [ 7] raxmlHPC-HYBRID[0x404bb6]
[comet-07-15:16651] [ 8] /lib64/libc.so.6(__libc_start_main+0xfd)[0x34d3c1ed5d]
[comet-07-15:16651] [ 9] raxmlHPC-HYBRID[0x402cc9]
[comet-07-15:16651] *** End of error message ***
/projects/ps-ngbt/home/cipres/ngbw/contrib/tools/bin/raxmlHPC-HYBRID_8.2.8: line 22: 16651 Aborted $EXE -T ${CIPRES_THREADSPP} $*
srun: error: comet-07-15: task 3: Exited with exit code 134
srun: First task exited 30s ago
srun: tasks 0-2,4-11: running
srun: task 3: exited abnormally
srun: Terminating job step 3680583.0
slurmstepd: *** STEP 3680583.0 ON comet-07-15 CANCELLED AT 2016-08-08T22:09:57 ***
srun: Job step aborted: Waiting up to 32 seconds for job step to finish.
/projects/ps-ngbt/backend/comet_workspace/NGBW-JOB-RAXMLHPC8_XSEDE-65AC62A225F54CBCBB0E0E4EAC636415/./_batch_command.cmdline: line 2: 16410 Killed raxmlHPC-HYBRID_8.2.8 -f a -N autoMRE -n result -s infile.txt -c 25 -p 12345 -m GTRCATI -x 12345
/projects/ps-ngbt/backend/comet_workspace/NGBW-JOB-RAXMLHPC8_XSEDE-65AC62A225F54CBCBB0E0E4EAC636415/./_batch_command.cmdline: line 2: 16409 Killed raxmlHPC-HYBRID_8.2.8 -f a -N autoMRE -n result -s infile.txt -c 25 -p 12345 -m GTRCATI -x 12345
srun: error: comet-07-15: tasks 0-1: Exited with exit code 137
srun: error: comet-07-15: tasks 2,4-5: Killed
srun: error: comet-07-20: tasks 6-11: Killed
raxmlHPC-HYBRID -T 4 -f a -N autoMRE -n result -s infile.txt -p 12345 -m GTRGAMMAI -x 12345Likelihood problem in model optimization l1: -inf l2: -8942163.8185108844190835952758789062500000000000 tolerance: 0.0000089421638185108840595795298700210196
#######
raxmlHPC-HYBRID: optimizeModel.c:3340: checkTolerance: Assertion `0' failed.
[comet-13-49:05788] *** Process received signal ***
[comet-13-49:05788] Signal: Aborted (6)
[comet-13-49:05788] Signal code: (-6)
[comet-13-49:05788] [ 0] /lib64/libpthread.so.0[0x3273e0f7e0]
[comet-13-49:05788] [ 1] /lib64/libc.so.6(gsignal+0x35)[0x3273232625]
[comet-13-49:05788] [ 2] /lib64/libc.so.6(abort+0x175)[0x3273233e05]
[comet-13-49:05788] [ 3] /lib64/libc.so.6[0x327322b74e]
[comet-13-49:05788] [ 4] /lib64/libc.so.6(__assert_perror_fail+0x0)[0x327322b810]
[comet-13-49:05788] [ 5] raxmlHPC-HYBRID[0x436e7e]
[comet-13-49:05788] [ 6] raxmlHPC-HYBRID[0x4473bd]
[comet-13-49:05788] [ 7] raxmlHPC-HYBRID[0x404bb6]
[comet-13-49:05788] [ 8] /lib64/libc.so.6(__libc_start_main+0xfd)[0x327321ed5d]
[comet-13-49:05788] [ 9] raxmlHPC-HYBRID[0x402cc9]
[comet-13-49:05788] *** End of error message ***
/projects/ps-ngbt/home/cipres/ngbw/contrib/tools/bin/raxmlHPC-HYBRID_8.2.8: line 22: 5788 Aborted $EXE -T ${CIPRES_THREADSPP} $*
srun: error: comet-13-49: task 3: Exited with exit code 134
raxmlHPC-HYBRID: optimizeModel.c:123: setRateModel: Assertion `rate >= 0.0001 && rate <= 1000000.0' failed.
It seems the sequences in the large alignment are correct, we succeeded in generating bootstrap trees in a separate command.
Thanks for your help,
Frederik
Alexandros Stamatakis
Sep 12, 2016, 2:51:47 AM9/12/16
to ra...@googlegroups.com
Dear Frederik,
Could you please try with the latest version RAxML v 8.2.9
https://github.com/stamatak/standard-RAxML/releases
If the problem still persists then please send me the input dataset to
my personal email.
All the best,
Alexis
On 09.09.2016 06:38, fsc...@berkeley.edu wrote:
> Dear Alexis,
>
> We are trying to run a large tree with 17492 nucleic acid sequences and
> 1533 alignment patterns using raxmlHPC-HYBRID-SSE3 (8.2.7) compiled (gcc
> version 4.8.5) on our cluster (each node has 256 Gb memory) with
> openmpi/1.8.7:
> |
> mpirun --np 20--map-bynode
> $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p 12345-N
> 20-m GTRCATI -n $suffix -C -T 16
> 12345-m GTRCATI -x 12345
> raxmlHPC-HYBRID -T 4-f a -N autoMRE -n result -s infile.txt -p 12345-m
> --
> 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
Could you please try with the latest version RAxML v 8.2.9
https://github.com/stamatak/standard-RAxML/releases
If the problem still persists then please send me the input dataset to
my personal email.
All the best,
Alexis
On 09.09.2016 06:38, fsc...@berkeley.edu wrote:
> Dear Alexis,
>
> We are trying to run a large tree with 17492 nucleic acid sequences and
> 1533 alignment patterns using raxmlHPC-HYBRID-SSE3 (8.2.7) compiled (gcc
> version 4.8.5) on our cluster (each node has 256 Gb memory) with
> openmpi/1.8.7:
> |
> mpirun --np 20--map-bynode
> $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p 12345-N
> |
>
> It runs without any issues on a smaller data set (~2000 sequences).
> However, with the large data set it crashes during final model optimization:
>
> Conducting final model optimizations on all 20 trees under GAMMA-based
> models ....
> #######
> raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion `rate
>>= 0.0001 && rate <= 1000000.0' failed.
>
>
> We also tested the "-F" option:
>
> Base frequencies: 0.254 0.230 0.313 0.203
> #######
> Likelihood problem in model optimization l1: -inf l2:
> -8438342.1678280774503946304321289062500000000000 tolerance:
> 0.0000084383421678280775418690140554645041
> --------------------------------------------------------------------------
> mpirun noticed that process rank 5 with PID 0 on node
> mc0208-ib.nersc.gov exited on signal 6 (Aborted).
> --------------------------------------------------------------------------
>
>
> And on the Cipres portal:
>
> |
> raxmlHPC-HYBRID -T 4-f a -N autoMRE -n result -s infile.txt -c 25-p
>
> It runs without any issues on a smaller data set (~2000 sequences).
> However, with the large data set it crashes during final model optimization:
>
> Conducting final model optimizations on all 20 trees under GAMMA-based
> models ....
> #######
> raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion `rate
>>= 0.0001 && rate <= 1000000.0' failed.
>
>
> We also tested the "-F" option:
>
> Base frequencies: 0.254 0.230 0.313 0.203
> #######
> Likelihood problem in model optimization l1: -inf l2:
> -8438342.1678280774503946304321289062500000000000 tolerance:
> 0.0000084383421678280775418690140554645041
> --------------------------------------------------------------------------
> mpirun noticed that process rank 5 with PID 0 on node
> mc0208-ib.nersc.gov exited on signal 6 (Aborted).
> --------------------------------------------------------------------------
>
>
> And on the Cipres portal:
>
> |
> 12345-m GTRCATI -x 12345
> 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
yann
Nov 6, 2017, 3:23:28 AM11/6/17
to raxml
Dear Alexis,
Similarly to Frederik, we are trying to run a large tree using raxmlHPC-AVX.PTHREADS (v8.2):
raxmlHPC -T 4 -s mintegra_seqs4raxml.phy -N autoMRE -n RAxML_t1.result
-g ttcons_phyla4RAXML_V2.tree -k -f a -m GTRCAT -x 12345 -p 12345 -o Arch_prof
However, RAxML spits the following error message and quits during the BS model parameter optimization:
raxmlHPC: optimizeModel.c:123: setRateModel: Assertion `rate >= RATE_MIN && rate <= RATE_MAX' failed.
zsh: abort (core dumped) raxmlHPC -T 4 -s mintegra_seqs4raxml.phy -N autoMRE -n RAxML_t1.result -g -k
We got similar error running the job on CIPRES:
[comet-12-53:32243] [ 1] /lib64/libc.so.6(gsignal+0x35)[0x3f3b432495]
[comet-12-53:32243] [ 2] /lib64/libc.so.6(abort+0x175)[0x3f3b433c75]
[comet-12-53:32243] [ 3] /lib64/libc.so.6[0x3f3b42b60e]
[comet-12-53:32243] [ 4] /lib64/libc.so.6(__assert_perror_fail+0x0)[0x3f3b42b6d0]
[comet-12-53:32243] [ 5] raxmlHPC-HYBRID[0x43d80d]
[comet-12-53:32243] [ 6] raxmlHPC-HYBRID[0x43a94e]
[comet-12-53:32243] [ 7] raxmlHPC-HYBRID[0x437ffa]
[comet-12-53:32243] [ 8] raxmlHPC-HYBRID[0x447b61]
[comet-12-53:32243] [ 9] raxmlHPC-HYBRID[0x404c4d]
[comet-12-53:32243] [10] /lib64/libc.so.6(__libc_start_main+0xfd)[0x3f3b41ed1d]
[comet-12-53:32243] [11] raxmlHPC-HYBRID[0x402d59]
[comet-12-53:32243] *** End of error message ***
/projects/ps-ngbt/home/cipres/ngbw/contrib/tools/bin/raxmlHPC-HYBRID_8.2.10_comet: line 22: 32243 Aborted $EXE -T ${CIPRES_THREADSPP} $*
srun: error: comet-12-53: task 7: Exited with exit code 134
raxmlHPC-HYBRID: optimizeModel.c:123: setRateModel: Assertion `rate >= 0.0001 && rate <= 1000000.0' failed.
[comet-12-53:32242] *** Process received signal ***
[comet-12-53:32242] Signal: Aborted (6)
[comet-12-53:32242] Signal code: (-6)
[comet-12-53:32242] [ 0] /lib64/libpthread.so.0[0x3f3bc0f7e0]
[comet-12-53:32242] [ 1] /lib64/libc.so.6(gsignal+0x35)[0x3f3b432495]
[comet-12-53:32242] [ 2] /lib64/libc.so.6(abort+0x175)[0x3f3b433c75]
[comet-12-53:32242] [ 3] /lib64/libc.so.6[0x3f3b42b60e]
[comet-12-53:32242] [ 4] /lib64/libc.so.6(__assert_perror_fail+0x0)[0x3f3b42b6d0]
[comet-12-53:32242] [ 5] raxmlHPC-HYBRID[0x43d80d]
[comet-12-53:32242] [ 6] raxmlHPC-HYBRID[0x43a94e]
[comet-12-53:32242] [ 7] raxmlHPC-HYBRID[0x437ffa]
[comet-12-53:32242] [ 8] raxmlHPC-HYBRID[0x447b61]
[comet-12-53:32242] [ 9] raxmlHPC-HYBRID[0x404c4d]
[comet-12-53:32242] [10] /lib64/libc.so.6(__libc_start_main+0xfd)[0x3f3b41ed1d]
[comet-12-53:32242] [11] raxmlHPC-HYBRID[0x402d59]
[comet-12-53:32242] *** End of error message ***
/projects/ps-ngbt/home/cipres/ngbw/contrib/tools/bin/raxmlHPC-HYBRID_8.2.10_comet: line 22: 32242 Aborted $EXE -T ${CIPRES_THREADSPP} $*
srun: error: comet-12-53: task 8: Exited with exit code 134
srun: First task exited 30s ago
srun: tasks 0,2-6,9-11: running
srun: tasks 1,7-8: exited abnormally
srun: Terminating job step 12455157.0
slurmstepd: *** STEP 12455157.0 ON comet-12-47 CANCELLED AT 2017-11-05T01:11:37 ***
srun: Job step aborted: Waiting up to 302 seconds for job step to finish.
/projects/ps-ngbt/backend/comet_workspace/NGBW-JOB-RAXMLHPC8_XSEDE-825F4951286D48298E904B1A07E60EA2/./_batch_command.cmdline: line 3: 12636 Killed raxmlHPC-HYBRID_8.2.10_comet -f a -N autoMRE -n result1 -s infile.txt -c 25 -m GTRCAT -p 785753221 -k -g constraint.tre -x 613658897 -o Arch_prof
srun: error: comet-12-47: tasks 0,2-5: Killed
srun: error: comet-12-53: tasks 6,9-11: KilledInterestingly, this error does not happen with a smaller data set or using the earlier versions of RAxML (i.e v.7) on the whole dataset
Any help would be welcome
Any help would be welcome
All the best,
Alexandros Stamatakis
Nov 6, 2017, 4:06:20 AM11/6/17
to ra...@googlegroups.com
Dear Yann,
Which version of RAxML are you using exactly?
This is indeed a dataset-dependent problem based on the error message
you received that is associated with model parameter optimization.
Did you try if this works with RAxML-NG the new, completely re-designed
version of RAxML?
https://github.com/amkozlov/raxml-ng
Alexis
On 06.11.2017 09:23, yann wrote:
> Dear Alexis,
>
> Similarly to Frederik, we are trying to run a large tree using
> raxmlHPC-AVX.PTHREADS (v8.2):
>
> *raxmlHPC -T 4 -s mintegra_seqs4raxml.phy -N autoMRE -n RAxML_t1.result
> 12345-m GTRCATI -x 12345
> raxmlHPC-HYBRID -T 4-f a -N autoMRE -n result -s infile.txt -p
> 12345-m GTRGAMMAI -x 12345
Which version of RAxML are you using exactly?
This is indeed a dataset-dependent problem based on the error message
you received that is associated with model parameter optimization.
Did you try if this works with RAxML-NG the new, completely re-designed
version of RAxML?
https://github.com/amkozlov/raxml-ng
Alexis
On 06.11.2017 09:23, yann wrote:
> Dear Alexis,
>
> Similarly to Frederik, we are trying to run a large tree using
> raxmlHPC-AVX.PTHREADS (v8.2):
>
> -g ttcons_phyla4RAXML_V2.tree -k -f a -m GTRCAT -x 12345 -p 12345 -o
> Arch_prof*
>
> However, RAxML spits the following error message and quits during the BS
> model parameter optimization:
>
> *raxmlHPC: optimizeModel.c:123: setRateModel: Assertion `rate >=
> However, RAxML spits the following error message and quits during the BS
> model parameter optimization:
>
> RATE_MIN && rate <= RATE_MAX' failed.
> zsh: abort (core dumped) raxmlHPC -T 4 -s mintegra_seqs4raxml.phy -N
> autoMRE -n RAxML_t1.result -g -k*
> zsh: abort (core dumped) raxmlHPC -T 4 -s mintegra_seqs4raxml.phy -N
> mpirun --np 20--map-bynode
> $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p
> 12345-N 20-m GTRCATI -n $suffix -C -T 16
> $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p
> |
>
> It runs without any issues on a smaller data set (~2000 sequences).
> However, with the large data set it crashes during final model
> optimization:
>
> Conducting final model optimizations on all 20 trees under
> GAMMA-based models ....
> #######
> raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion
> `rate >= 0.0001 && rate <= 1000000.0' failed.
>
>
> We also tested the "-F" option:
>
> Base frequencies: 0.254 0.230 0.313 0.203
> #######
> Likelihood problem in model optimization l1: -inf l2:
> -8438342.1678280774503946304321289062500000000000 tolerance:
> 0.0000084383421678280775418690140554645041
> --------------------------------------------------------------------------
> mpirun noticed that process rank 5 with PID 0 on node
> mc0208-ib.nersc.gov <http://mc0208-ib.nersc.gov> exited on signal 6
>
> It runs without any issues on a smaller data set (~2000 sequences).
> However, with the large data set it crashes during final model
> optimization:
>
> Conducting final model optimizations on all 20 trees under
> GAMMA-based models ....
> #######
> raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion
> `rate >= 0.0001 && rate <= 1000000.0' failed.
>
>
> We also tested the "-F" option:
>
> Base frequencies: 0.254 0.230 0.313 0.203
> #######
> Likelihood problem in model optimization l1: -inf l2:
> -8438342.1678280774503946304321289062500000000000 tolerance:
> 0.0000084383421678280775418690140554645041
> --------------------------------------------------------------------------
> mpirun noticed that process rank 5 with PID 0 on node
> (Aborted).
> --------------------------------------------------------------------------
>
>
> And on the Cipres portal:
>
> |
> raxmlHPC-HYBRID -T 4-f a -N autoMRE -n result -s infile.txt -c 25-p
> --------------------------------------------------------------------------
>
>
> And on the Cipres portal:
>
> |
> 12345-m GTRCATI -x 12345
> 12345-m GTRGAMMAI -x 12345
> --
> 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
> 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
yann
Nov 6, 2017, 4:16:58 AM11/6/17
to raxml
Dear Alexis,
Many thanks for your quick reply.
I am using the Standard RAxML version 8.2.11 on my computer and the 8.2.9 on CIPRES.
I did not try yet to run the RAxML-NG... Is it available on CIPRES?
Best,
Many thanks for your quick reply.
I am using the Standard RAxML version 8.2.11 on my computer and the 8.2.9 on CIPRES.
I did not try yet to run the RAxML-NG... Is it available on CIPRES?
Best,
Alexandros Stamatakis
Nov 6, 2017, 4:23:44 AM11/6/17
to ra...@googlegroups.com
Okay, and you get the same error in both cases, correct?
RAxML-NG is not yet available on CIPRES.
Could you send me the input dataset to my personal email such that I c
an have a look?
Thanks,
Alexis
On 06.11.2017 10:16, yann wrote:
> Dear Alexis,
>
> Many thanks for your quick reply.
>
> I am using the Standard RAxML version 8.2.11 on my computer and the
> 8.2.9 on CIPRES.
>
> I did not try yet to run the RAxML-NG... Is it available on CIPRES?
>
> Best,
>
> Yannick
>
>
> On Friday, September 9, 2016 at 6:38:13 AM UTC+2, fsc...@berkeley.edu wrote:
>
> Dear Alexis,
>
> We are trying to run a large tree with 17492 nucleic acid sequences
> and 1533 alignment patterns using raxmlHPC-HYBRID-SSE3 (8.2.7)
> compiled (gcc version 4.8.5) on our cluster (each node has 256 Gb
> memory) with openmpi/1.8.7:
> |
> 12345-m GTRCATI -x 12345
> raxmlHPC-HYBRID -T 4-f a -N autoMRE -n result -s infile.txt -p
> 12345-m GTRGAMMAI -x 12345
RAxML-NG is not yet available on CIPRES.
Could you send me the input dataset to my personal email such that I c
an have a look?
Thanks,
Alexis
On 06.11.2017 10:16, yann wrote:
> Dear Alexis,
>
> Many thanks for your quick reply.
>
> I am using the Standard RAxML version 8.2.11 on my computer and the
> 8.2.9 on CIPRES.
>
> I did not try yet to run the RAxML-NG... Is it available on CIPRES?
>
> Best,
>
> Yannick
>
>
> On Friday, September 9, 2016 at 6:38:13 AM UTC+2, fsc...@berkeley.edu wrote:
>
> Dear Alexis,
>
> We are trying to run a large tree with 17492 nucleic acid sequences
> and 1533 alignment patterns using raxmlHPC-HYBRID-SSE3 (8.2.7)
> compiled (gcc version 4.8.5) on our cluster (each node has 256 Gb
> memory) with openmpi/1.8.7:
> |
> mpirun --np 20--map-bynode
> $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p
> 12345-N 20-m GTRCATI -n $suffix -C -T 16
> $HOME/apps/standard-RAxML/bin/raxmlHPC-HYBRID-SSE3 -s $align -p
> |
>
> It runs without any issues on a smaller data set (~2000 sequences).
> However, with the large data set it crashes during final model
> optimization:
>
> Conducting final model optimizations on all 20 trees under
> GAMMA-based models ....
> #######
> raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion
> `rate >= 0.0001 && rate <= 1000000.0' failed.
>
>
> We also tested the "-F" option:
>
> Base frequencies: 0.254 0.230 0.313 0.203
> #######
> Likelihood problem in model optimization l1: -inf l2:
> -8438342.1678280774503946304321289062500000000000 tolerance:
> 0.0000084383421678280775418690140554645041
> --------------------------------------------------------------------------
> mpirun noticed that process rank 5 with PID 0 on node
> mc0208-ib.nersc.gov <http://mc0208-ib.nersc.gov> exited on signal 6
>
> It runs without any issues on a smaller data set (~2000 sequences).
> However, with the large data set it crashes during final model
> optimization:
>
> Conducting final model optimizations on all 20 trees under
> GAMMA-based models ....
> #######
> raxmlHPC-HYBRID-SSE3: optimizeModel.c:123: setRateModel: Assertion
> `rate >= 0.0001 && rate <= 1000000.0' failed.
>
>
> We also tested the "-F" option:
>
> Base frequencies: 0.254 0.230 0.313 0.203
> #######
> Likelihood problem in model optimization l1: -inf l2:
> -8438342.1678280774503946304321289062500000000000 tolerance:
> 0.0000084383421678280775418690140554645041
> --------------------------------------------------------------------------
> mpirun noticed that process rank 5 with PID 0 on node
> (Aborted).
> --------------------------------------------------------------------------
>
>
> And on the Cipres portal:
>
> |
> raxmlHPC-HYBRID -T 4-f a -N autoMRE -n result -s infile.txt -c 25-p
> --------------------------------------------------------------------------
>
>
> And on the Cipres portal:
>
> |
> 12345-m GTRCATI -x 12345
> 12345-m GTRGAMMAI -x 12345
yann
Nov 6, 2017, 5:39:05 AM11/6/17
to raxml
Email sent. Thank you for your help
Best,
yannick
Best,
yannick
haris zaf
Apr 24, 2018, 10:12:59 AM4/24/18
to raxml
Dear Alexis,
I have a similar issue.
I ran:
/usr/bin/raxmlHPC -f a -p 12345 -s consensus_sequences.fasta -x 54321 -# 100 -m GTRGAMMA -n WITH_Ts
and I got the .TEST i am sending you and a .sh.eXXXX file with this error:
raxmlHPC-PTHREADS-AVX: optimizeModel.c:123: setRateModel: Assertion `rate >= RATE_MIN && rate <= RATE_MAX' failed.
Aborted
I use RAxML version 8.2.9. Could you help me out somehow please?
Thank you very much!
Haris
I have a similar issue.
I ran:
/usr/bin/raxmlHPC -f a -p 12345 -s consensus_sequences.fasta -x 54321 -# 100 -m GTRGAMMA -n WITH_Ts
and I got the .TEST i am sending you and a .sh.eXXXX file with this error:
raxmlHPC-PTHREADS-AVX: optimizeModel.c:123: setRateModel: Assertion `rate >= RATE_MIN && rate <= RATE_MAX' failed.
Aborted
I use RAxML version 8.2.9. Could you help me out somehow please?
Thank you very much!
Haris
Alexandros Stamatakis
Apr 24, 2018, 10:34:19 AM4/24/18
to ra...@googlegroups.com
Dear Haris,
Please try RAxML-NG and see if that works on your dataset:
https://github.com/amkozlov/raxml-ng
as support of the old standard RAxML will be progressively reduced.
Alexis
Please try RAxML-NG and see if that works on your dataset:
https://github.com/amkozlov/raxml-ng
as support of the old standard RAxML will be progressively reduced.
Alexis
haris zaf
Apr 24, 2018, 12:13:08 PM4/24/18
to ra...@googlegroups.com
Τhank you very much Alexi!!
I think that it does work!! On Tue, Apr 24, 2018 at 5:34 PM, Alexandros Stamatakis <alexandros...@gmail.com> wrote:
Dear Haris,
Please try RAxML-NG and see if that works on your dataset:
https://github.com/amkozlov/raxml-ng
as support of the old standard RAxML will be progressively reduced.
Alexis
On 24.04.2018 16:12, haris zaf wrote:
Dear Alexis,
I have a similar issue.
I ran:
/usr/bin/raxmlHPC -f a -p 12345 -s consensus_sequences.fasta -x 54321 -# 100 -m GTRGAMMA -n WITH_Ts
and I got the .TEST i am sending you and a .sh.eXXXX file with this error:
raxmlHPC-PTHREADS-AVX: optimizeModel.c:123: setRateModel: Assertion `rate >= RATE_MIN && rate <= RATE_MAX' failed.
Aborted
I use RAxML version 8.2.9. Could you help me out somehow please?
Thank you very much!
Haris
--
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+unsubscribe@googlegroups.com <mailto:raxml+unsubscribe@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
--
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+unsubscribe@googlegroups.com.
taua...@gmail.com
Jan 22, 2019, 10:23:20 AM1/22/19
to raxml
Dear Alexis,
I believe I am having the same problem, but from the previous messages couldn't figure it out what to do.
I am running raxml on over 1000 gene alignments. All of them worked, except one, that keeps failing with the error message below.
I prefer to fix it for this one, instead of having to install and redo all 1000 gene trees again. Do you have any recommendations?
The alignment itself looks fine. I am using RAxML 8.2.10
Thank you very much!
Best,
Tauana
raxmlHPC-HYBRID-SSE3: rapidBootstrap.c:432: treeOptimizeThorough: Assertion `res > 0' failed.
[holygiribet03:26751] *** Process received signal ***
[holygiribet03:26751] Signal: Aborted (6)
[holygiribet03:26751] Signal code: (-6)
[holygiribet03:26751] [ 0] /lib64/libpthread.so.0(+0xf5e0)[0x2ab88bfc35e0]
[holygiribet03:26751] [ 1] /lib64/libc.so.6(gsignal+0x37)[0x2ab88c2091f7]
[holygiribet03:26751] [ 2] /lib64/libc.so.6(abort+0x148)[0x2ab88c20a8e8]
[holygiribet03:26751] [ 3] /lib64/libc.so.6(+0x2e266)[0x2ab88c202266]
[holygiribet03:26751] [ 4] /lib64/libc.so.6(+0x2e312)[0x2ab88c202312]
[holygiribet03:26751] [ 5] /n/holylfs/LABS/giribet_lab/tauanajc/scripts/RAxML/raxmlHPC-HYBRID-SSE3[0x484b09]
[holygiribet03:26751] [ 6] /n/holylfs/LABS/giribet_lab/tauanajc/scripts/RAxML/raxmlHPC-HYBRID-SSE3[0x43c4a5]
[holygiribet03:26751] [ 7] /n/holylfs/LABS/giribet_lab/tauanajc/scripts/RAxML/raxmlHPC-HYBRID-SSE3[0x425f5f]
[holygiribet03:26751] [ 8] /lib64/libc.so.6(__libc_start_main+0xf5)[0x2ab88c1f5c05]
[holygiribet03:26751] [ 9] /n/holylfs/LABS/giribet_lab/tauanajc/scripts/RAxML/raxmlHPC-HYBRID-SSE3[0x4069dd]
[holygiribet03:26751] *** End of error message ***
srun: error: holygiribet03: task 1: Aborted (core dumped)
On Tuesday, April 24, 2018 at 12:13:08 PM UTC-4, haris wrote:
HarisThanks again!Τhank you very much Alexi!!I think that it does work!!
On Tue, Apr 24, 2018 at 5:34 PM, Alexandros Stamatakis <alexandros...@gmail.com> wrote:
Dear Haris,
Please try RAxML-NG and see if that works on your dataset:
https://github.com/amkozlov/raxml-ng
as support of the old standard RAxML will be progressively reduced.
Alexis
On 24.04.2018 16:12, haris zaf wrote:
Dear Alexis,
I have a similar issue.
I ran:
/usr/bin/raxmlHPC -f a -p 12345 -s consensus_sequences.fasta -x 54321 -# 100 -m GTRGAMMA -n WITH_Ts
and I got the .TEST i am sending you and a .sh.eXXXX file with this error:
raxmlHPC-PTHREADS-AVX: optimizeModel.c:123: setRateModel: Assertion `rate >= RATE_MIN && rate <= RATE_MAX' failed.
Aborted
I use RAxML version 8.2.9. Could you help me out somehow please?
Thank you very much!
Haris
--
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
--
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.
Alexandros Stamatakis
Jan 23, 2019, 1:01:57 AM1/23/19
to ra...@googlegroups.com
Dear Tauana,
I'd also want to ask you to use RAxML-NG:
https://github.com/amkozlov/raxml-ng
The error you get indicates that there is a numerical issue with the
code which I think, should not occur in RAxML-NG.
Also, if you have to infer gene trees on many MSAs, please check this
new convenient tool we developed:
https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty839/5132696
Alexis
> <javascript:> <mailto:raxml+un...@googlegroups.com
> <javascript:>>.
I'd also want to ask you to use RAxML-NG:
https://github.com/amkozlov/raxml-ng
The error you get indicates that there is a numerical issue with the
code which I think, should not occur in RAxML-NG.
Also, if you have to infer gene trees on many MSAs, please check this
new convenient tool we developed:
https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty839/5132696
Alexis
> <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>
>
> --
> Alexandros (Alexis) Stamatakis
>
> Research Group Leader, Heidelberg Institute for Theoretical Studies
> Full Professor, Dept. of Informatics, Karlsruhe Institute of
> Technology
>
>
>
> --
> 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 <javascript:>.
>
> --
> You received this message because you are subscribed to the
> Google Groups "raxml" group.
> To unsubscribe from this group and stop receiving emails from
> For more options, visit https://groups.google.com/d/optout
> <https://groups.google.com/d/optout>.
taua...@gmail.com
Jan 23, 2019, 1:15:16 PM1/23/19
to raxml
Thanks Alexis, I was hoping avoid installing the NG right now because it was just one alignment, but installing was easy and it ran very fast.
I think the problem happened the same, but the error message in RAxML-NG was much more informative (there was one amino acid missing from that alignment, and I was using empirical frequencies). I simply used the matrix frequencies and then it ran properly.
Thanks!
Tauana
Alexandros Stamatakis
Jan 23, 2019, 2:47:59 PM1/23/19
to ra...@googlegroups.com
great :-)
alexis
On 23.01.19 19:15, taua...@gmail.com wrote:
> Thanks Alexis, I was hoping avoid installing the NG right now because it
> was just one alignment, but installing was easy and it ran very fast.
>
> I think the problem happened the same, but the error message in RAxML-NG
> was much more informative (there was one amino acid missing from that
> alignment, and I was using empirical frequencies). I simply used the
> matrix frequencies and then it ran properly.
>
> Thanks!
> Tauana
>
> On Wednesday, January 23, 2019 at 1:01:57 AM UTC-5, Alexis wrote:
>
> Dear Tauana,
>
> I'd also want to ask you to use RAxML-NG:
> https://github.com/amkozlov/raxml-ng
> <https://github.com/amkozlov/raxml-ng>
>
> The error you get indicates that there is a numerical issue with the
> code which I think, should not occur in RAxML-NG.
>
> Also, if you have to infer gene trees on many MSAs, please check this
> new convenient tool we developed:
>
> https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty839/5132696
> <https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty839/5132696>
>
>
> Alexis
>
alexis
On 23.01.19 19:15, taua...@gmail.com wrote:
> Thanks Alexis, I was hoping avoid installing the NG right now because it
> was just one alignment, but installing was easy and it ran very fast.
>
> I think the problem happened the same, but the error message in RAxML-NG
> was much more informative (there was one amino acid missing from that
> alignment, and I was using empirical frequencies). I simply used the
> matrix frequencies and then it ran properly.
>
> Thanks!
> Tauana
>
> On Wednesday, January 23, 2019 at 1:01:57 AM UTC-5, Alexis wrote:
>
> Dear Tauana,
>
> I'd also want to ask you to use RAxML-NG:
> https://github.com/amkozlov/raxml-ng
> <https://github.com/amkozlov/raxml-ng>
>
> The error you get indicates that there is a numerical issue with the
> code which I think, should not occur in RAxML-NG.
>
> Also, if you have to infer gene trees on many MSAs, please check this
> new convenient tool we developed:
>
> https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty839/5132696
> <https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty839/5132696>
>
>
> Alexis
>
Reply all
Reply to author
Forward
0 new messages