From: Jeff Comer (jeffcomer_at_gmail.com)
Date: Thu Feb 26 2015 - 17:34:09 CST
This depends on what's in your model. As far as I know, the general
consensus is that flexible covalent bonds to hydrogen oscillate to
quickly to use a 2 fs timestep without substantial error. I've heard
of people increasing the mass of hydrogen atoms to get around this
problem, but this can, of course, alter the kinetics of the system.
United-atom models are another option.
Note that with many popular force fields (e.g. CHARMM), using
"rigidBonds none" is a bad idea.
These force fields were parameterized with rigid covalent bonds
involving hydrogen and rigid water. Using "rigidBonds none" with these
force fields means that you are not implementing the force field
correctly. It's not a small error either. Using the fallback flexible
parameters for CHARMM TIP3P gives you a water dielectric constant of
175.6 ± 1.7 (according to my calculations), which is about twice what
I get with rigid water.
Jeffrey Comer, PhD
Institute of Computational Comparative Medicine
Nanotechnology Innovation Center of Kansas State
Kansas State University
Office: P-213 Mosier Hall
On Thu, Feb 26, 2015 at 4:49 PM, Hasan haska
> Dear All,
> I have some questions to ask about configuration file in namd. Is it right
> to set "timestep 2 fs" with "rigidBonds none" in a configuration file when
> running simulations ? Does it negatively affect the correctness of the
> simulation ? And If I choose "rigidBonds none" in NVT, can I write
> "langevinHydrogen off” in the configuration file ?
> Thanks for the reply.
This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:20:56 CST