Thanks for your explanation. I realized that I might only be able to use
gb=1 to do minimization and nmode() finally. I asked the question to
mailing list because I was thinking, what if... what if somebody has come
up with a way to solve that that I just don't know...
For the spacing parameters, I mean the "space" option in mm_options().
According to manual, it sets the grid spacing for the finite difference
solver. Honestly, I didn't fully understand its usage. But I read from Dr.
Ray Luo's answer to other users who had the same problems: enlarge space to
see if helps. In that case, space was increased to 1.25 to finally solve
the PB Bomb. But it never worked in my case...
Thank you for your time again!
2014-05-09 15:47 GMT-04:00 David A Case <case.biomaps.rutgers.edu>:
> On Fri, May 09, 2014, Guqin Shi wrote:
> >
> > I was using the XMIN method in NAB package to minimize my
> > complex(protein+small molecule). The reason I am doing that is I want to
> > further calculate entropy by using nmode() in NAB.
> >
> > I was trying to use PB solvent in minimization and below is my input:
>
> This won't work, for at least two reasons:
>
> 1. The only implicit solvent model that has second derivatives (needed
> for nmode()) is igb=1.
>
> 2. The xmin minimizer has strong assumptions about the accuracy of the
> forces;
> numerical procdures like PB or RISM don't satisfy these conditions. So you
> would not likely be able to do a full minimization anyway.
>
> Finally, your particular allocation error suggests that you are asking for
> more memory than you have available. You don't give enough details about
> "changing the spacing parameters" to be very helpful (at least to me).
>
> ...dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
>
--
Guqin SHI
The Ohio State University
College of Pharmacy
608 Riffe Building
496 W. 12th Ave.
Columbus, OH, 43210
(614)688-3531
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri May 09 2014 - 14:00:03 PDT
