VMD-L Mailing List
From: Stefano Guglielmo (stefano.guglielmo_at_unito.it)
Date: Fri Feb 03 2017 - 02:01:33 CST
thanks for clarification, I was too strict in expecting the same values for
energy and colvars, but actually it's true, they are almost the same thing.
2017-02-03 4:21 GMT+01:00 Giacomo Fiorin <giacomo.fiorin_at_gmail.com>:
> Hi Stefano, if I sort the contents of the PMF file according to the energy
> I get that the lowest free-energy values are:
> 2.5 115 -0
> 3.5 125 0.0138202
> 3.5 115 0.23787
> If I sort the last part of the "output" file, and look for the largest
> values of the metadynamics potential I get:
> 2985 2.65351685319544e+00 1.14605320428208e+02 11.4118
> 2987 2.70255164664564e+00 1.12710621783704e+02 11.4129
> 3737 3.74336457789873e+00 1.21202158456548e+02 11.4188
> So, again, the putative free-energy minimum is in the neighborhood of
> distance = 3 Å, angle = 120°. Clearly the PMF has a grid with round values
> for the two variables and the trajectory does not, but that's the only
> difference that I can see.
> On Thu, Feb 2, 2017 at 8:04 PM, Stefano Guglielmo <
> stefano.guglielmo_at_unito.it> wrote:
>> Hi Giacomo,
>> to be honest, I am missing the use of "cv save" to write a pmf file...
>> besides this, I had a look at the file, generated in vmd, reporting
>> metadynamics potential for each frame (and the corresponding values of the
>> two colvars I defined) and I found the inconsistency I wrote in the
>> previous post: the "configuration" corresponding to the 0 in the pmf
>> generated in NAMD is not the one with the largest value of metadynamics
>> potential; moreover I could not find the numerical relationship between
>> metadynamics potential and pmf: does this behavior depend on somewhat wrong
>> choice of colvars or bias parameters?
>> The two files are attached with the second reporting frames
>> distance angle energy.
>> 2017-02-01 15:43 GMT+01:00 Giacomo Fiorin <giacomo.fiorin_at_gmail.com>:
>>> Hi Stefano, besides the obvious issue that the metadynamics potential
>>> converges to the negative of the PMF, can you try using the cv save
>>> command in VMD, to write a PMF file?
>>> This file could then be compared to the one written by NAMD (which you
>>> already have), and detect any inconsistencies.
>>> On Tue, Jan 31, 2017 at 9:02 PM, Stefano Guglielmo <
>>> stefano.guglielmo_at_unito.it> wrote:
>>>> Hi all,
>>>> For sure I am missing something but I've got a bit confused about the
>>>> discrepancy between the pmf minimum obtained from a metadynamics simulation
>>>> (ie the colvars values corresponding to 0) and the minimum value of
>>>> metadynamics potential calculated for the same simulation through vmd
>>>> colvars module, corresponding to quite different values of colvars compared
>>>> to the previous ones.
>>>> Can anybody give some hints or references?
>>>> Thanks in advance
>>>> Stefano GUGLIELMO PhD
>>>> Assistant Professor of Medicinal Chemistry
>>>> Department of Drug Science and Technology
>>>> Via P. Giuria 9
>>>> 10125 Turin, ITALY
>>>> ph. +39 (0)11 6707678
>> Stefano GUGLIELMO PhD
>> Assistant Professor of Medicinal Chemistry
>> Department of Drug Science and Technology
>> Via P. Giuria 9
>> 10125 Turin, ITALY
>> ph. +39 (0)11 6707678
-- Stefano GUGLIELMO PhD Assistant Professor of Medicinal Chemistry Department of Drug Science and Technology Via P. Giuria 9 10125 Turin, ITALY ph. +39 (0)11 6707678