RE: OPLS dihedrals

From: JC Gumbart (gumbart_at_ks.uiuc.edu)
Date: Sat May 19 2007 - 11:35:08 CDT

You would put the following:

A B C D K1 n1 d1
A B C D K2 n2 d2

And so on. The Charmm force field also has some sets of atoms with multiple
dihedral terms so it is not unreasonable.

In case you haven't found it yet, take a look at
http://www.ks.uiuc.edu/Research/namd/wiki/index.cgi?NamdAndOPLS

-----Original Message-----
From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf
Of Arturas Ziemys
Sent: Saturday, May 19, 2007 10:47 AM
To: NAMD-LIST
Subject: namd-l: OPLS dihedrals

Hi,

I want to simulate long n-alkanes with NAMD using OPLS-UA FF (Jorgensen,
1984). But I miss something how to put dihedral Fourier parameters into NAMD
"style". For OPLS there are optimized 3 dihedral sinusoids, if I'm following
right (V = V0 + 1/2V1(1+cos(w)) + 1/2V2(1-cos(2w)) + 1/2V3(1+cos(3w))).

NAMD has for dihedral K (1 + cos( n*(w-d) )). Does it mean that I should put
tree sinusoids into NAMD parameter file like :

A B C D K n d K n d K n d
?

How to adopts Jorgensen dihedral potential to NAMD ? Or, in general, is
there any sense to include that dihedral potential into MD of long n-alkanes
?

Arturas

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:19 CST