Re: SMD at constant velocity

From: Giacomo Fiorin (giacomo.fiorin_at_gmail.com)
Date: Tue Dec 28 2021 - 10:06:14 CST

Impossible to say for sure without digging deeper into the details of your
system, but there are a couple of common things to check.

1. Make sure that PBC wrapping is not an issue for your variables'
definition:
https://urldefense.com/v3/__https://colvars.github.io/colvars-refman-namd/colvars-refman-namd.html*sec:colvar_atom_groups_wrapping__;Iw!!DZ3fjg!tzT0yseWc4Avjl1Bm1eoiGGpw8eq0pRlRRfVwyQjM4z2JdJN-OYyETDC_fqIEfoFKA$
and test them individually.

2. Try an unbiased MD run first, i.e. defining just the variables but not
the restraint to get an idea of how the initial force will act.

Giacomo

On Tue, Dec 28, 2021 at 5:45 AM Almeida-Hern√°ndez, Yasser, Dr. <
yasser.almeida-hernandez_at_uni-due.de> wrote:

> Dear Giacomo,
>
>
> Thank you for your reply.
>
>
> I've tried with different initial and final centers, up to 10 A apart,
> and the same happens. The proteins reach the targetCenters very fast and
> stay there, while the centers evolve linearly until the end of the
> simulation.
>
>
> What do you think could be the problem?
>
>
> Best,
>
> Yasser
> ------------------------------
> *From:* Giacomo Fiorin <giacomo.fiorin_at_gmail.com>
> *Sent:* Monday, December 27, 2021 5:14:57 PM
> *To:* NAMD list; Almeida-Hern√°ndez, Yasser, Dr.
> *Subject:* Re: namd-l: SMD at constant velocity
>
> Hi Yasser, it sounds like the variables' values do not evolve as you
> expect them to, but the restraint centers do work as expected. This is not
> uncommon.
>
> Your initial and final centers differ by only 0.5 √Ö, and assuming you used
> the default width of 1, for a force constant of 5 kcal/mol you
> get 1/2√ó5√ó0.5^2 = 0.625 kcal/mol, roughly one kT unit. Not much difference
> between those states as far as the restraint is concerned.
>
> Accumulated work, like all energies, is in the same unit of the MD engine
> that Colvars is distributed with, in this case NAMD:
>
> https://urldefense.com/v3/__https://colvars.github.io/colvars-refman-namd/colvars-refman-namd.html*sec:units__;Iw!!DZ3fjg!tzT0yseWc4Avjl1Bm1eoiGGpw8eq0pRlRRfVwyQjM4z2JdJN-OYyETDC_fqwltreLw$
> and NAMD uses kcal/mol.
>
> Giacomo
>
> On Mon, Dec 27, 2021 at 5:47 AM Almeida-Hern√°ndez, Yasser, Dr. <
> yasser.almeida-hernandez_at_uni-due.de> wrote:
>
>> Dear all,
>>
>>
>> I am struggling with a SMD at constant velocity, where I want to pull two
>> proteins continuously along the X-axis apart, at k = 5 kcal/mol at 0.5
>> A/ns, during 50 ns
>>
>>
>> Below is the definition of the harmonic:
>>
>>
>> harmonic {
>> colvars pull_nter pull_cter pull_nter_y pull_cter_y
>> pull_nter_z pull_cter_z
>> centers -20.0 20.0 0.0 0.0
>> 0.0 0.0
>> targetCenters -20.5 20.5 0.0 0.0 0.0
>> 0.0
>> forceConstant 5.
>> targetNumSteps 25000000
>> outputCenters yes
>> outputAccumulatedWork yes
>> }
>>
>>
>> I have tried several centers and targetCenters to achieve the velocity,
>> but what happens is that the targetCenters are reached very fast and
>> then the proteins stay there. However, I write the centers during the
>> simulation (with outputCenters, x0_pull_nter and x0_pull_nter) and they
>> change continuously as expected until the end of the run.
>>
>>
>> Am I missing something?
>>
>>
>> Also, what are the units of the accumulated work written in the .traj
>> file?
>>
>>
>> Any help will be very much appreciated.
>>
>>
>> Best,
>>
>> Yasser
>>
>>
>> --
>> Dr. Yasser Almeida-Hernandez
>> Postdoctoral Researcher
>> Computational Biochemistry group
>> T01 R03 D45
>> Faculty of Biology
>> University of Duisburg-Essen
>> Universitätsstr. 2, 45117 Essen
>> Email: yasser.almeida-hernandez_at_uni-due.de
>> <%20yasser.almeida-hernandez_at_uni-due.de>
>> Phone: +49 201 183 2457 <+49%20201%20183%202457>
>>
>>

This archive was generated by hypermail 2.1.6 : Fri Dec 31 2021 - 23:17:12 CST