Hi Harut
Yes, it works and we have been using it for rescoring virtual Hits for
some time now. I would also suggest adding 1000-5000 steps of minimisation
before MM/PBSA.
Since, MMPBSA is very sensitive to internal dielectric constant, you may
want to look at it carefully for other proteins where it fails.
On Tuesday, November 3, 2020, <h_sahakyan.mb.sci.am> wrote:
> Dear Dr. Case,
>
> I understand that MM/GBSA is designed to calculate energies using MD
> snapshots. Nevertheless, I would like to use it for docking results
> rescoring and any recommendations will be very much appreciated.
> The initial benchmarks demonstrated that it works and improves AUC and
> Enrichment factor, though in the case of some proteins it fails.
> Interestingly, I noticed that minimization using just 500-1000 steps yields
> better results than full minimization (with drms = 1.0E-4 for example).
> What do you think about it?
>
> Best wishes,
> Harut
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Best Regards
Elvis
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 02 2020 - 18:00:02 PST
