Maybe I have not made my point clear.
AMBER force field is OK. I tried combination of 03, 99SB and IGB5, IGB1. All of them are good.
But PESs from MP2 and FF are different. I reparametrized some ff terms according to MP2, and then I got large deviation from exp, with large population of left-handed alpha helix.
So I guess the AMBER FF is not purely derived from MP2 calculation, but with some manual modifications. If yes, I would like to know how the developers changed the PES. Maybe I can follow them.
Ye
2012-01-13
From£º Daniel Roe
Date£º 2012-01-13 21:15:47
To£º AMBER Mailing List
CC£º
Subject£º Re: [AMBER] PES for dialanine
Hi,
What solvent model are you using? In simulations I have done and have seen
for various polyalanines (at least up to decaalanine) with a reasonable
solvent model and FF99SB, the most populated secondary structure type is
PP2, which agrees with expt.
On Thursday, January 12, 2012, Ye MEI <ymei.itcs.ecnu.edu.cn> wrote:
> Dear Amber community,
>
> Recently we plot the 2D potential energy surface of di-alanine spanned by
main chain torsions (phi and psi). The structures are optimized at
HF/6-31G* level, and the single point energies are calcualted at
MP2/aug-ccpvtz level. And It shows large difference from that of AMBER,
either 03 or 99SB. I put the PES on the web, you can download them from
> http://itcs.ecnu.edu.cn/ymei/03_total_PES.pdf
> http://itcs.ecnu.edu.cn/ymei/99SB_total_PES.pdf
> http://itcs.ecnu.edu.cn/ymei/MP2_total.pdf
>
> The main difference is located at phi around 70 degree, and psi between
-150 and 50 degree. Has Amber force field been manually optimized,
especially to reduce the population of left-handed alpha helix? If I use
the MP2 PES in MD simulation of triAlanine, I get large population of
left-handed alpha helix, which seems to be against the experiment.
>
> Any comments or suggestions are welcomed.
>
> Ye MEI
> 2012-01-13
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
-------------------------
Daniel R. Roe, PhD
Postdoctoral Associate
BioMaPS Institute, Rutgers University
610 Taylor Road
Piscataway, NJ 08854
_______________________________________________
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 Jan 13 2012 - 06:30:04 PST
