Re: question about FEP soft-core

From: Floris Buelens (floris_buelens_at_yahoo.com)
Date: Tue Sep 15 2009 - 08:56:43 CDT

The soft core vdW potential is still 12-6 - the exponents are 6 and 3 but they are applied to r^2 rather than r. If you plug in lambda=1 or delta=0 you recover standard Lennard-Jones.
Yes, using the parameters you suggest is the same as running without a soft core potential, but this is inadvisable in virtually all circumstances so make sure you have a very good reason for doing so.
Best,

Floris

________________________________
From: yun luo <luoyun724_at_gmail.com>
To: NAMD <namd-l_at_ks.uiuc.edu>
Sent: Monday, 14 September, 2009 21:03:16
Subject: namd-l: question about FEP soft-core

Hi,

I have a few question about the soft-core in NAMD. Does it include three parameters as below or just fepVdwShiftCoeff ?
fepVdwShiftCoeff (delta in eq.28);
fepElecLambdaStart (related to Lambda(elec) in eq.28);
fepVdwLambdaEnd (related to Lambda(LJ) in eq.28)

The Vdw part in eq.28 use 6-3 LJ potential instead of typical 12-6 LJ potential. Is it because the soft-core only apply to short-range Vdw interactions (repulsive part of LJ potential)?
What's the major difference between NAMD soft-core and the WCA repulsive pair potential used in CHARMM?

If I don't want use soft-core, should I set the three parameters as below?
epVdwShiftCoeff 0 (default value 5)
fepElecLambdaStart 0 (default value 0.5)
fepVdwLambdaEnd 1.0 (default value 1.0)

Thanks a lot!

Lyna

      

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:22:23 CST