[gmx-users] Re: Error There is no domain decomposition for 6 nodes that is compatible

[Justin Lemkul]

On 10/5/12 5:19 PM, Sonia Aguilera wrote:
> Thank you a lot,
>
> I only changed the couple-intramol setting (couple-intramol=yes) and now
> it´s running just fine. However I have a doubt about something. In the
> manual says the following when using couple-intramol=no
>
> "In this manner the decoupled state of the molecule corresponds to the
> proper vacuum state without periodicity effects"
>
> I don´t understand the real effects and implications of this on my
> simulation. Does it mean that it is better to run with couple-intramol=no? I
> also read that using couple-intramol=yes has it's advantages: "This can be
> useful for partitioning free-energies of relatively large molecules, where
> the intra-molecular non-bonded interactions might lead to kinetically
> trapped vacuum conformations. Again, I don't understand what is the real
> meaning of that. Can you please make this clear for me?
>

Consider what (de)coupling means.  You are manipulating the interactions of a 
chosen molecule with its surroundings as a function of lambda.  If you tell 
mdrun that your calculation should not couple intramolecular interactions 
(couple-intramol = no) then what that is saying is that all nonbonded 
interactions within that given molecule are always calculated at full strength. 
  That is what the manual is telling you - the calculation implies that 
intramolecular terms are always on, as if the molecule were in a vacuum.  If you 
do couple intramolecular interactions (couple-intramol = yes), then any 
nonbonded interactions within your molecule of choice are also scaled as a 
function of lambda.  This may be beneficial for large molecules, since if you 
are not coupling intramolecular interactions, you may get unnaturally strong 
interactions within the solute molecule since the interactions with the 
surrounding solvent are weakened as a function of lambda.  Thus you can get odd 
geometries that get trapped and are detrimental to your sampling.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================