PBMetaD + multi-walkers restart broken in GMX2020.5 + latest plumed

189 views
Skip to first unread message

Jim Pfaendtner

unread,
Mar 3, 2021, 5:45:31 PM3/3/21
to plumed...@googlegroups.com
Hello everyone,

I have just compiled the latest version available on the plumed github along with GMX2020.5.  Multiwalkers and PBMetaD seems to be broken for restarting jobs. When a job is restarted the replicas > 0 do not get the correct hill height. Instead, their hill heights are reset to the starting values. 

An example HILLS file is below showing a 5 ps run , then restarting after that time. My input files are attached. 

We’ve confirmed this 3X independently in my group so it seems real :) 

Thanks for any advice!

Jim


#! FIELDS time d sigma_d height biasf
#! SET multivariate false
#! SET kerneltype gaussian
                    0.5     0.4194357060693539                   0.02     0.4799999999999999                      6
                    0.5     0.4194357060693539                   0.02     0.4799999999999999                      6
                    0.5     0.4194348383278976                   0.02     0.4799999999999999                      6
                    0.5     0.4194357060693539                   0.02     0.4799999999999999                      6
                    0.5     0.4194357060693539                   0.02     0.4799999999999999                      6
                      1     0.3827485040719882                   0.02     0.4927441169520531                      6
                      1     0.3827485040719882                   0.02     0.4927441169520531                      6
                      1     0.3827580075494151                   0.02     0.4927120627387296                      6
                      1     0.3827485040719882                   0.02     0.4927441169520531                      6
                      1     0.3827485040719882                   0.02     0.4927441169520531                      6
..
..
..

                      5     0.3622347238499357                   0.02     0.5670991707466346                      6
                      5     0.3622347238499357                   0.02     0.5670991707466346                      6
                      5     0.3906584832737987                   0.02     0.1203538172460751                      6
                      5     0.3622347238499357                   0.02     0.5670991707466346                      6
                      5     0.3622347238499357                   0.02     0.5670991707466346                      6
#! FIELDS time d sigma_d height biasf
#! SET multivariate false
#! SET kerneltype gaussian
                    5.5     0.3558429144210522                   0.02     0.2426687475648539                      6
                    5.5     0.3502892838041164                   0.02     0.4799999999999999                      6
                    5.5     0.3873635721593541                   0.02     0.4799999999999999                      6
                    5.5     0.3502892838041164                   0.02     0.4799999999999999                      6
                    5.5     0.3502892838041164                   0.02     0.4799999999999999                      6


broke_pb.tgz

Giovanni Bussi

unread,
Mar 3, 2021, 5:58:06 PM3/3/21
to plumed...@googlegroups.com

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/B35CC188-812A-4801-97F9-4D299DBF89EB%40gmail.com.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/B35CC188-812A-4801-97F9-4D299DBF89EB%40gmail.com.

Jim Pfaendtner

unread,
Mar 3, 2021, 6:05:41 PM3/3/21
to plumed...@googlegroups.com
Thanks Giovanni!

This fixes it for PB. It is strange because I did not need this directive for WTM and so I initially missed it when I was helping diagnose the problem. Below you can see a HILLS file from a WTM simulation. That HILLS file is in the rep0 directory even though I am running 5 replicas (1.44 initial height). 

Probably as best practice it is useful to just specify ../HILLSx so there is no question. 

Best,
Jim


                    100     -1.594949477125556      2.589715770267064                    0.2                    0.2     0.3282636794440898                      6
                    100     -1.773975558845275     -2.616595128059225                    0.2                    0.2      1.087560351719508                      6
                    100    -0.7308928427925121     -0.191002418299977                    0.2                    0.2     0.7723349399133237                      6
                    100     -1.581906058217572     0.6543131866278067                    0.2                    0.2     0.2835328330745652                      6
                    100    -0.9641057704873135    -0.2803032809727455                    0.2                    0.2     0.5161409953296647                      6
#! FIELDS time phi psi sigma_phi sigma_psi height biasf
#! SET multivariate false
#! SET kerneltype gaussian
#! SET min_phi -pi
#! SET max_phi pi
#! SET min_psi -pi
#! SET max_psi pi
                  100.5      2.993715596276434      2.771734419800762                    0.2                    0.2     0.9541965592005012                      6
                  100.5     -2.802480853797258      2.139380664814551                    0.2                    0.2     0.4799725672752547                      6
                  100.5     -2.475330850082436      2.753437517394638                    0.2                    0.2     0.2153965822737904                      6
                  100.5     -1.532304228007514     0.8368412749146206                    0.2                    0.2     0.2358539638416754                      6
                  100.5     -1.269289363206339      1.275652427253855                    0.2                    0.2     0.2421782103620935                      6


Billy Noonan

unread,
May 12, 2021, 12:03:46 AM5/12/21
to PLUMED users
Dear Jim and Giovanni,

Sorry to bring up an old thread, but any help with the following would be much appreciated...
https://groups.google.com/g/plumed-users/c/RiVveeYZbN8

Best Wishes,
Billy
Reply all
Reply to author
Forward
0 new messages