Re: Metadynamics simulation

From: Giacomo Fiorin (giacomo.fiorin_at_gmail.com)
Date: Thu Jan 05 2017 - 09:51:08 CST

At the moment NAMD does not have a "turn-key" implementation for
bias-exchange metadynamics neither through Colvars (which is already
included in NAMD) or through Plumed (which could be added separately,
although it is primarily coupled with GROMACS).

Note, however, that NAMD has a Tcl scripting interface that is capable of
sending commands between replicas:
*http://www.ks.uiuc.edu/Research/namd/2.12/ug/node9.html#SECTION00052300000000000000
<http://www.ks.uiuc.edu/Research/namd/2.12/ug/node9.html#SECTION00052300000000000000>*
and this can be used in conjunction with Colvars' own Tcl scripting
interface via the cv command:
*http://colvars.github.io/colvars-refman-namd/colvars-refman-namd.html#x1-820007
<http://colvars.github.io/colvars-refman-namd/colvars-refman-namd.html#x1-820007>*
Basically, you need to use different configuration files for each bias, and
do exchanges by loading/unloading them:
cv bias *oldbiasname* delete
cv configfile <*newbias configuration*>
The energy for the Metropolis criterion can be obtained at any time through:
cv bias *biasname* energy
So, with NAMD and Colvars you are required to do a little scripting (feel
free to adapt the files in lib/replica) but then you are able to customize
the algorithm much more.

As for multiple-walkers metadynamics (i.e. no exchanges, but data are
continuously shared between replicas), it is available in NAMD+Colvars out
of the box:
*http://colvars.github.io/colvars-refman-namd/colvars-refman-namd.html#x1-690006.3.4
<http://colvars.github.io/colvars-refman-namd/colvars-refman-namd.html#x1-690006.3.4>*

Giacomo

On Thu, Jan 5, 2017 at 7:39 AM, Azadeh Alavi <azade.alavi_at_gmail.com> wrote:

>
> Hello Giacomo
>
> Thanks a lot for your explanations. I upgraded to NAMD2.12 and the problem
> was solved,and as your said, the performance also improved.
>
> One more question is that is it possible to do bias-exchange metadynamics
> with NAMD and Plumed or one should turn to GROMACS?
>
> Thank you very much for your time and help.
>
> Best,
> Azadeh
>
> On Wed, Jan 4, 2017 at 9:46 PM, Giacomo Fiorin <giacomo.fiorin_at_gmail.com>
> wrote:
>
>> Hello Azadeh
>>
>> On Tue, Jan 3, 2017 at 10:00 AM, Azadeh Alavi <azade.alavi_at_gmail.com>
>> wrote:
>>
>>> Dear NAMD users,
>>>
>>> I am going to perform Metadynamics simulations for ion transport through
>>> a protein. I am using NAMD2.11. I have copied the input file below.
>>> First, I was wondering if a timestep of 2 fs is OK to run this
>>> simulation and whether the parameters below are reasonable.
>>>
>>
>> To have any chance of giving meaningful results, the metadynamics forces
>> should be much smaller than the ones generated from the force field, so the
>> same criteria of the unbiased simulation would apply for the time step.
>>
>>
>>> Secondly, when I run this simulation, the output file of HillsTrajectory
>>> is empty which is weird.
>>>
>>
>> This a glitch caused by a redesign of the file i/o interface for NAMD
>> prior to the release of NAMD 2.11. The hills are being added, and you
>> should be able to see them in the state file. Also, you should notice an
>> applied force in the colvars trajectory.
>>
>> Please upgrade to NAMD 2.12. You should also gain performance
>> improvements because certain variables such as distanceZ are now computed
>> fully in parallel.
>>
>> Giacomo
>>
>>
>>
>>>
>>> Thanks a lot for your help.
>>> Azadeh
>>>
>>>
>>>
>>> colvar {
>>> name Translocation
>>> lowerBoundary -4.0
>>> lowerWallConstant 100.0
>>> upperBoundary 56.0
>>> upperWallConstant 100.0
>>> width 0.4
>>> outputVelocity on
>>> outputAppliedForce on
>>>
>>> distanceZ {
>>> main {
>>> atomnumbers { 131678 }
>>> }
>>> ref {
>>> atomnumbers { protein alpha carbons..
>>> }
>>> }
>>> axis ( 0.0, 0.0, 1.0 )
>>> }
>>> }
>>>
>>> metadynamics {
>>>
>>> colvars Translocation
>>> hillWeight 0.01 # Default is 0.01 - higher numbers = faster
>>> newHillFrequency 100 # Default is 100 - lower numbers = faster
>>> writeFreeEnergyFile on
>>> saveFreeEnergyFile on
>>> keephills on
>>> writeHillsTrajectory on
>>> }
>>>
>>
>>
>>
>> --
>> Giacomo Fiorin
>> Associate Professor of Research
>> Institute for Computational Molecular Science (ICMS)
>> College of Science and Technology, Temple University
>> 1925 North 12th Street (035-07), Room 704D
>> Philadelphia, PA 19122-1801
>> Phone: +1-215-204-4213 <(215)%20204-4213>
>>
>> Scholar: http://goo.gl/Q3TBQU
>> Personal: http://giacomofiorin.github.io/
>> Lab page: https://icms.cst.temple.edu/members.html
>>
>> *"As computer programmers we have a responsibility to make sure that we
>> run the computers instead of the computers running us."* - Steve
>> Oualline
>>
>>
>>
>

-- 
Giacomo Fiorin
Associate Professor of Research
Institute for Computational Molecular Science (ICMS)
College of Science and Technology, Temple University
1925 North 12th Street (035-07), Room 704D
Philadelphia, PA 19122-1801
Phone: +1-215-204-4213 <(215)%20204-4213>
Scholar: http://goo.gl/Q3TBQU
Personal: http://giacomofiorin.github.io/
Lab page: https://icms.cst.temple.edu/members.html
*"As computer programmers we have a responsibility to make sure that we run
the computers instead of the computers running us."*  - Steve Oualline

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:57 CST