[gmx-users] Free energy calculation, parallel or sequential lambda runs ?

[Stéphane Téletchéa]

David Mobley a écrit :
> Stéphane,
> 
> 
>> I'm willing to do some FEP using the slow groth approach available in
>> gromacs (as of 3.3.1). While retrieving articles and the mailing list, i
>> found out there where arguments against sequential runs (particularly
>> for avoiding the hysteresis problem), and other against parallel runs.
> 
> If you are doing slow growth, that is an entirely different thing from
> doing parallel or sequential lambda runs. Slow growth gradually
> changes lambda from 0 to 1 in a continuous manner over the course of a
> simulation and I don't recommend it unless you have a particular
> problem where this is the only way to do it or something. But
> generally it leads to Hamiltonian lag.

You are right, i was confusing both terms, of course i want to do 
parallels runs, thus specifying each lambda.
This has both advantages of avoiding the Hamiltonian lag and saving 
computer time (since i can run the 20 simulations in parallel instead of 
serial).

Thanks a lot for your long and clear explanation about procedure and 
time, it'll save some time for getting experience in the field :-)

I'll try to update my construction procedure (to be published soon) to 
take your advices into account.

Best wishes,

Stéphane

-- 
Stéphane Téletchéa, PhD.                  http://www.steletch.org
Unité Mathématique Informatique et Génome http://migale.jouy.inra.fr/mig
INRA, Domaine de Vilvert                  Tél : (33) 134 652 891
78352 Jouy-en-Josas cedex, France         Fax : (33) 134 652 901