[gmx-users] Doubt about Free Energy control Minimization

[Justin Lemkul]

On 7/6/17 8:42 AM, Varvdekar Bhagyesh Rajendra wrote:
> Dear Justin,
> 
> It would be helpful if you please explain in a bit more detail about the impact of the settings of table-extension. The manual says "Extension of the non-bonded potential lookup tables beyond the largest cut-off distance", but a better insight from you would be definitely more very useful.
> 
> I have a slightest idea that it is useful to save computational time. It that is considered true, it shouldn't affect the results, should it?
> 
> I would appreciate if you could explain how increasing table-extension can cause permit distorted structures.
> 
> I performed a short study of the effect on increasing table-extension from 1 nm to 10 nm onwards in increments of 1 nm. The inter and intra molecular energies  increased from 1 nm to 4 nm and then were almost the same and with the minimization without any free energy code. Hence, I suspect that table-extension with 4 nm onwards takes into account all interactions while the free energy change goes on. I would appreciate if you could offer any comment on my deduction.
> 

If you increase the table-lookup length, you tell mdrun that it is OK to find 
things like 1-4 interactions at some very long distance.  Normally when a 1-4 
goes beyond 1 nm, it is a sign of trouble.  If you make it normal, you turn a 
blind eye to possible problems.

> I couldn't find any distorted structures while the study though, could you suggest any criteria so as to consider if any structure has evolved into a distorted one?
> 
> Is there any other way where I can avoid the said warning other than increasing table-extension and maintaining couple-intramol = no in my minimizing code with the free energy addendum.
> 

No idea, sorry.  And unfortunately it appears that no one else on the mailing 
list has any idea about what's going on in your system, either.  Maybe someone 
else will chip in...

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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