RESTART METADYNAMICS

[Gianvito Grasso]

Dear PLUMED users,
I'm using the metadynamics with GROMACS 5.1.2+plumed-2.2.3

I need to restart a metadynamics simulation. In order to do this, I added RESTART at the beginning of the plumed.dat  file, together with GRID_RFILE, as reported below: 

"

RESTART

WHOLEMOLECULES STRIDE=1 ENTITY0=1-7654

metad: METAD ARG=d PACE=1000 HEIGHT=0.4 SIGMA=0.03 FILE=HILLS BIASFACTOR=20.0 TEMP=300.0 GRID_MIN=3.0 GRID_MAX=7.0 GRID_BIN=500 GRID_WSTRIDE=1000 GRID_WFILE=GRID GRID_RFILE=GRID

"

However, I obtain the following COLVAR file after the restart:

 

18994.000902 4.305455 0.248041 78.554668

 18996.000902 4.327314 0.227650 81.986054

 18998.000902 4.298810 0.230917 78.081839

 19000.000902 4.316786 0.232985 80.441169

#! FIELDS time d t metad.bias

#! SET min_t -pi

#! SET max_t pi

 19000.000902 4.316786 0.232985 160.585943

 19002.000903 4.322839 0.234386 162.632556

 19004.000903 4.315838 0.216161 160.468182

 19006.000903 4.299419 0.152854 156.221650

 

For some reasons, the metad.bias seems to be doubled after the restart point.

How can I avoid this problem? Any help on this is appreciated.

Thank you very much.

All the Best,

Gianvito

[Massimiliano Bonomi]

Hello,

Are you reading from both the HILLS file and the grid written on file?

Max

Sent from my Nexus. Please excuse typos and brevity.

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to plumed...@googlegroups.com.
Visit this group at https://groups.google.com/group/plumed-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/a952eb4e-1e66-49c7-a7ed-c4323bb34acc%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Gianvito Grasso]

Dear Max,

thank you for your quick reply.

From what I understood (I don't know if it is correct), the GRID_RFILE=GRID action allows to read the bias deposited on a grid file.

What do you mean when you said "reading from HILLS"?

Anyway, I used several times the same command line in plumed.dat with a different plumed version (plumed 2.3b+gromacs-5.1.2) without any problems.

Thank you very much for your help

Gianvito

To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.

[Massimiliano Bonomi]

Normally when you add the RESTART keyword you read the Gaussians written on the HILLS file to calculate the initial bias potentials. 

When you are restarting, both the HILLS file from the previous run and the grid are present in the current directory? Looks like you readong both

Sent from my Nexus. Please excuse typos and brevity.

To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users+unsubscribe@googlegroups.com.

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

To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/6013b904-9b7f-482a-befa-5085d1c9a736%40googlegroups.com.

[Gianvito Grasso]

Ok now it's clear! Yes, both HILLS and GRID are present in the current directory.

So, can I restart the metadynamics with GRID_RFILE=GRID without RESTART keyword, in order to read only from grid?

Thanks!

GG

[Massimiliano Bonomi]

But in this way you will not append your COLVAR files. Alternatively i think you can move the HILLS file from that directory and store it somewhere else. I am not sure what is the common practice in these situations, since I don't use the write/read to grid usually, but I typically restart from the HILLS files.

Sent from my Nexus. Please excuse typos and brevity.

To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users+unsubscribe@googlegroups.com.

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

To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/5831e89e-7c3c-422c-9c0c-039304202262%40googlegroups.com.

[Gianvito Grasso]

Ok, thank you very much!

All the best

Gianvito