REST2 metadynamics test with Gromacs 4.6.7 and PLUMED v2.1-hrex -> no output

993 views
Skip to first unread message

peter.rich...@gmail.com

unread,
Dec 1, 2014, 6:22:59 AM12/1/14
to plumed...@googlegroups.com

I used an alanine dipeptide in TIP3P test system. I scaled the solute as described in the Hrex tutorial:

https://github.com/GiovanniBussi/plumed2/blob/v2.1-hrex/user-doc/tutorials/hrex.txt

Analog to Giovannis Hrex paper:

http://www.tandfonline.com/doi/abs/10.1080/00268976.2013.824126#.VHxHPa0S0Yp


Is the following sentence in the tutorial outdated?

'Also notice that only the patch for gromacs 4.6.5 implementes hrex.'


Because, a pure metadynamics and a pure REST work, but as long as I apply both it seems that the simulation stuck at the first exchange without an Error and without stopping.


I start with:

mpirun -np 6 mdrun_mpi -v -s topol -plumed plumed.dat -multi 2 -replex 200 -nsteps 20000 -hrex


Inputs and exemplary outputs of the 'stucked run' are attached.


Why does the simulation stuck at the first exchange? Am I missing an important flag?


Thanks a lot in advance.


Cheers

Peter

COLVAR.0
grompp0.mdp
md0.log
plumed.dat

Giovanni Bussi

unread,
Dec 1, 2014, 6:36:39 AM12/1/14
to plumed...@googlegroups.com
Hi!

On Mon, Dec 1, 2014 at 12:22 PM, <peter.rich...@gmail.com> wrote:

I used an alanine dipeptide in TIP3P test system. I scaled the solute as described in the Hrex tutorial:

https://github.com/GiovanniBussi/plumed2/blob/v2.1-hrex/user-doc/tutorials/hrex.txt

Analog to Giovannis Hrex paper:

http://www.tandfonline.com/doi/abs/10.1080/00268976.2013.824126#.VHxHPa0S0Yp


Is the following sentence in the tutorial outdated?

'Also notice that only the patch for gromacs 4.6.5 implementes hrex.'


This should be updated to "4.6.7". I am likely not going to port this to gromacs 5.0

 


Because, a pure metadynamics and a pure REST work, but as long as I apply both it seems that the simulation stuck at the first exchange without an Error and without stopping.


I start with:

mpirun -np 6 mdrun_mpi -v -s topol -plumed plumed.dat -multi 2 -replex 200 -nsteps 20000 -hrex


Inputs and exemplary outputs of the 'stucked run' are attached.


Why does the simulation stuck at the first exchange? Am I missing an important flag?


It looks you are doing the right thing. Just to be sure, if you run identical command line with an empty plumed.dat file, does the simulation get stuck? 

Giovanni


Thanks a lot in advance.


Cheers

Peter

--
You received this message because you are subscribed to the Google Groups "PLUMED users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.
To post to this group, send email to plumed...@googlegroups.com.
Visit this group at http://groups.google.com/group/plumed-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/6be780eb-a877-4fe6-931e-61aa235d8f6c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

peter.rich...@gmail.com

unread,
Dec 1, 2014, 9:00:07 AM12/1/14
to plumed...@googlegroups.com

Thanks, for the comment that I am basically doing the right thing :)



The identical simulation with an empty plumed.dat file runs fine. Furthermore, a Hrex simulation without the metadynamics runs as well, see attachment.

Is: Repl 0 <-> 1 dE_term = -0.000e+00 (kT) normal for this system and each exchange? See md0.log



To show my dead body in the basement, I configured PLUMED with CXX=mpicc and also CC=mpicc and have not compiled additional libraries like BLAS or LAPACK it run through I guess with the internal ones.


Cheers

Peter


md0.log
plumed.dat

peter.rich...@gmail.com

unread,
Dec 2, 2014, 8:53:16 AM12/2/14
to plumed...@googlegroups.com, peter.rich...@gmail.com

Excuse, to phrase a concrete question:


Can a mistake in the building/patching procedure cause the stuck behavior I observed? As long as they have completed without an error.


I would highly appreciate if someone with a running Gromacs-Hrex version could test my system to exclude that this is a system problem and reproducible. (Files attached, running command in the first post)


Thanks a lot!


Peter

plumed.dat
topol0.tpr
topol1.tpr

zhou...@gmail.com

unread,
Jan 6, 2015, 7:54:31 AM1/6/15
to plumed...@googlegroups.com, peter.rich...@gmail.com
Hi Peter:
Recently,I am also considering of running REST in GROMACS4.6.5. But I meet some problems. From what I understood,I did this by changing the ffnonbonded.itp as well as the ffbonded.itp in share/groamcs/top file to add the scale factor to the force field parameters and then using a script to generate a set of .top files for each replica.I did a short test with  alanine dipeptide using 3 replicas. However,I found that in the md.log file the  dE =  0.000e+00,and the exchange probability between two neighboring replicas was 1.0, I was so confused with this.so could you please give me some suggestions about this question? thanks in advance.
best wishes

在 2014年12月2日星期二UTC+8下午9时53分16秒,peter.rich...@gmail.com写道:

Giovanni Bussi

unread,
Jan 8, 2015, 4:27:27 AM1/8/15
to plumed...@googlegroups.com
Hi all!

You should not edit the itp. You should first produce a single top file (grompp -pp) then edit that file, as explained in the documentation. I guess this will solve the problem.

Regards,

Giovanni


--
You received this message because you are subscribed to the Google Groups "PLUMED users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.
To post to this group, send email to plumed...@googlegroups.com.
Visit this group at http://groups.google.com/group/plumed-users.

Giovanni Bussi

unread,
Jan 8, 2015, 4:33:39 AM1/8/15
to plumed...@googlegroups.com
Hi Peter,

I tried your input. Notes:
1. plumed.dat does not work on my machine because it containes ^M characters. Please convert it with dos2unix
2. even without plumed, your setup crashes on my machine after a few exchanges. Possibly you chose a too low lambda factor (which I cannot check given the tpr only). Anyway, simulation is not stuck at the exchange.

I hope this helps,

Giovanni

--
You received this message because you are subscribed to the Google Groups "PLUMED users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.
To post to this group, send email to plumed...@googlegroups.com.
Visit this group at http://groups.google.com/group/plumed-users.

周双艳

unread,
Jan 10, 2015, 10:01:31 PM1/10/15
to plumed...@googlegroups.com
Hi Giovanni:
Thanks for you kind reply. Following your suggestions, I changed my way to implement REST,
First,Using grompp -pp to get a processed.top file; second,Using a script to do the appropriate solute parameter scaling for each replica; And then,Using grompp normally on the new set of .top files to generate a set of .tpr files that differ not only in lambda but also in their solute parameters. In the .mdp file, the only difference between each replicas is the init-lambda-state and keep all other parameter the same.But the result just the same as before,the term "Repl 0 <-> 1 dE = 0.000e+00" in md.log file still with the value of 0 ,and the exchange probability of two neighboring replicas still with the value of 1.0
In order to explain this result ,I then calculated the potential energy distributions of replicas,I found that potential energy of each replica different with each other indeed but with little difference. And my groamcs4.6.5 install without PLUMED,is this will influence implementing H-REMD in gromacs? thanks for your reply again
Best wishes
Shuangyan Zhou

--
You received this message because you are subscribed to a topic in the Google Groups "PLUMED users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/plumed-users/rbbZxfHDrIE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to plumed-users...@googlegroups.com.

To post to this group, send email to plumed...@googlegroups.com.
Visit this group at http://groups.google.com/group/plumed-users.

Giovanni Bussi

unread,
Jan 16, 2015, 2:40:02 AM1/16/15
to plumed...@googlegroups.com
Hi,

I am not sure lambda stuff will work together with hrex stuff. Please choose either one or the other. Lambda stuff is probably better supported (it is in the official gromacs, no need to patch), but, in my experience, slower.

Giovanni


peter.rich...@gmail.com

unread,
Mar 27, 2015, 11:31:26 AM3/27/15
to plumed...@googlegroups.com
Solved!

The mentioned stucking problem is solved by using this post:
https://groups.google.com/forum/#!topic/plumed-users/h4AzwSMhRmE

Thanks Giovanni and Mengzhi han!!

All the best!
Peter
Reply all
Reply to author
Forward
0 new messages