I agree with Matias about the misconception, MD is not always faster in GB-
it can actually take more computer resources per time stepas discussed in
the links he provided . However it can parallelize better, and some global
motions may be sampled in many fewer steps than corresponding dynamics in
explicit solvent. My view is that these are two different models, each with
their own strengths and weaknesses. One is not "better" than the other. For
some projects one is the right choice, for other projects it is not. There
is always a tradeoff between precision and accuracy. One must be aware of
the limitations of all of our models and choose the appropriate one.
Another good example is QM vs MM - each is the appropriate tool in a
different situation.
Carlos
On Thu, Sep 12, 2019 at 10:38 AM Matias Machado <mmachado.pasteur.edu.uy>
wrote:
> Just a follow up comment to David's...
>
> There seems to be a general misconception about the speedup of implicit
> solvent simulations... most people believe they are faster that explicit
> ones just because there is no water molecules, but they usually neglect
> several other issues of implementation and so on...
>
> I recommend you to read these two previous threats on the topic:
>
> + http://archive.ambermd.org/201807/0309.html
> + http://archive.ambermd.org/201806/0279.html
>
> Another advice for free... sometime you do need explicit solvent
> (water/ions) to correctly describe an interaction or phenomenon and there
> is no workaround for that... so, if someone wants to sell you a Ferrari for
> a dollar, then be suspicious...
>
> Best,
>
> MatÃas
>
> ------------------------------------
> PhD.
> Researcher at Biomolecular Simulations Lab.
> Institut Pasteur de Montevideo | Uruguay
> [http://pasteur.uy/en/labs/biomolecular-simulations-laboratory]
> [http://www.sirahff.com]
>
> ----- Mensaje original -----
> De: "Kolattukudy P. Santo" <santotheophys.gmail.com>
> Para: "David Case" <david.case.rutgers.edu>, "AMBER Mailing List" <
> amber.ambermd.org>
> Enviados: Martes, 3 de Septiembre 2019 22:45:42
> Asunto: Re: [AMBER] GBSA
>
> Thank you, Dave for the answers
> Regards
> Santo
>
> On Tue, 3 Sep 2019 at 21:19, David Case <david.case.rutgers.edu> wrote:
>
> > On Tue, Sep 03, 2019, Kolattukudy P. Santo wrote:
> >
> > >I am trying to do implicit solvent simulations in AMBER, with
> generalized
> > >Born solvation. (1) Is it really faster than explicit MD? ( I dont find
> it
> > >is that faster !).
> >
> > It is often the case that explicit solvent simualtions (which take
> > advantage of PME to handle long-range electrostatic interactions) can be
> > faster than the corresponding implicit solvent simulations (which don't
> > make use of PME). Balancing off against this is that, generally,
> > conformational exploration is faster in implicit solvents, so you should
> > not just look at ns/day as the only metric.
> >
> >
> > (2) Is it appropriate for simulating very long polymers, say
> > >with 100 or more monomers ?
> >
> > Sure.....dac
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
>
> --
> *Dr. K. P. Santo*
> *Post doctoral Associate *
> *Department of Chemical and Biochemical Engineering*
> *Rutgers, The State University of New Jersey*
> *New Brunswick, New jersey*
> *USA*
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Sep 13 2019 - 06:30:06 PDT
