Re: Rapid and accurate parameterization of new molecules in CHARMM

From: Kenno Vanommeslaeghe (
Date: Wed May 28 2014 - 10:35:07 CDT

On 05/27/2014 08:22 PM, Aaron Larsen wrote:
> I'd like to evaluate many small hypothetical modifications.

Define "small". Sure, adding or subtracting a methyl or ethyl group to or
from an existing ring in the NA FF would qualify as small, but I'd be hard
pressed to come up with 100+ such modifications. Conversely, if you're
going to change elements or saturation/oxidation states in the ring, its
character changes completely and you're well outside the coverage of the
NA FF and in CGenFF territory.

Also, actual parameter optimization on 100+ molecules will presumably take
more than a month even when using one of the automated interfaces. I'd say
give a try. If some of the proposed molecules get warnings
or really bad penalties, you can still decide whether to run them through
ffTK or skip them.

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:25 CST