Umbrella sampling using PLUMED 2.3.0 in DL_POLY 4.08

[ustc...@gmail.com]

Hi all,

I hope this is the right place where I can find some help.

I installed PLUMED 2.3.0 with DL_POLY 4.08. Everything works without any errors when I run an umbrella sampling. However, the biased potential seems not having any effect. The simulation runs as if it does not have the biased potential on the fly.

I also tried extremely large KAPPA. COLVAR file shows the biased potential value as expected. But it does not affect the simulation result.

Any suggestions would be highly appreciated.

Thanks,

Yi

[Gareth Tribello]

Can you post your input for PLUMED and DL_POLY so we know more precisely what you are trying to do?

Thanks

Gareth

--
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 https://groups.google.com/group/plumed-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/bcaaa506-f512-4fc1-ba50-49220f675d2d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[ustc...@gmail.com]

Yes, Gareth. I attach a zip file here with 6 files in it.

It runs a molecular dynamics with umbrella sampling by adding a constraint with respect to volume V. I intend to set the V0 = 46.6663. And KAPPA is already ridiculously huge I think.

COLVAR shows that the biased potential goes pretty large.

Thanks,

Yi

[Giovanni Bussi]

Can you check a variable other than the volume?

Inviato da iPhone

To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/5b1c22cd-fab2-4c32-a9e7-9b582bc462e3%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

<temp.zip>

[Alin M Elena]

Hi Yi,

TEST28 here shall have an umbrella sampling test.
ftp://ftp.dl.ac.uk/ccp5/DL_POLY/DL_POLY_4.0/DATA/
as far as I am aware shall work.

Regards,
Alin

Without Questions there are no Answers!
_______________________________________
Dr. Alin Marin ELENA
http://alin.elena.space/
_______________________________________

[ustc...@gmail.com]

I am checking. It may take several hours and I will definitely let you know. Thanks!

[ustc...@gmail.com]

Hi Alin,

Thanks for the suggestion.

I ran TEST28 before. It was working well and I was sure that the biased potential had an effect on the fly. 

I think I am using pretty much similar input except that the CV is volume in my case.

Cheers,

Yi

[Massimiliano Bonomi]

Forgive my naive question.
Are you running your simulation in the NPT ensemble?

Max

> --
> 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 https://groups.google.com/group/plumed-users.

> To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/27cc964c-1f27-4079-9679-789eb49c1a9c%40googlegroups.com.

[ustc...@gmail.com]

Hi Max,

No. I am actually running NσT ensemble where σ is a stress tensor. This is different from NPT by allowing shape to change as well.

Cheers,

Yi

[Giovanni Bussi]

I suspect dlpoly does not receive the virial correctly from plumed. The plumed interface is embedded in dlpoly, so you should check this with dlpoly developers or with the person that implemented the interface.

In the plumed developer manual there are examples showing how to debug the communication of virial between plumed and the host md code

Inviato da iPhone

To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/ea490985-88f3-4978-9dd1-1727b3032364%40googlegroups.com.

[Massimiliano Bonomi]

It is very peculiar the time serie of the volume and bias CV.
You start very close to the value you are putting the umbrella at, and you
rapidly go far away from it, clearly against the bias potential.
You obtain “exactly” the opposite effect of an umbrella restraint.
Which let me think that maybe there is a wrong sign (on the DL_POLY end) in the
application of the forces. This might happens if the PLUMED derivatives are
interpreted as forces. Maybe is a good place to start when checking the DL_POLY interface.

Max

> To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/D6963561-12BA-4918-AE51-83CCAB94E012%40gmail.com.

[Alin M Elena]

Hi Giovanni and Max,

It may be indeed a bug.
this is the bit that deals with both virial and forces...

140 Call plumed_f_gcmd("setForcesX"//sn,fxx)
141 Call plumed_f_gcmd("setForcesY"//sn,fyy)
142 Call plumed_f_gcmd("setForcesZ"//sn,fzz)
143 plumed_virial = -stress
144 Call plumed_f_gcmd("setVirial"//sn,plumed_virial)
145 stress = -plumed_virial

I will have to look on the virial over the weekend. I wonder why I got
sensible results for buthane. I would have expect any of the problems you
mention to show already there.

Regards,
Alin

[Giovanni Bussi]

You have to do this test:

https://plumed.github.io/doc-master/developer-doc/html/_how_to_plumed_your_m_d.html

Check virial contribution

Let us know if it works!

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+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/6582086.A2MJLbFIFc%40circassia.dl.ac.uk.

[ustc...@gmail.com]

Hi Giovanni,

As I said yesterday, I did two other simulations to test. One is without any constraint. The other is constraining "by", the y component of b vector of the box. This variable is supposed to vary greatly without constraint.

However, both simulations behave as if there are not any constraints. So I think the volume is not simply an exception.

Cheers,

Yi

On Thursday, March 23, 2017 at 3:43:12 PM UTC-5, Giovanni Bussi wrote:

[ustc...@gmail.com]

Hi Giovanni,

I did the Check Virial Contribution. Two output files are identical. Still, I am using NσT ensemble.

Thanks,

Yi

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 https://groups.google.com/group/plumed-users.

[Matej Badin]

Hi Yi,

Can you please share your files for the Check Virial test from Developer documentation you mentioned. I run using it my files also in NσT (anisotropic Berendsen barostat) and I was not

able to obtain the same thing. Also were you successful at the end?

Best,

Matej