Re: Computational electrophysiology

From: Gumbart, JC (
Date: Tue Aug 06 2019 - 20:21:49 CDT

Take a look at this approach, which we demonstrated a few years back:


On Aug 5, 2019, at 8:56 AM, Giacomo Fiorin <<>> wrote:

Hello José, to my knowledge most simulations of TM proteins with NAMD have been done with a constant voltage difference across the unit cell and a single-bilayer setup.
Occasionally, the electric field may vary over the unit cell by using grifForces instead of eField, to mimic the differences in dielectric screening in the different compartments.

I'm not sure if there is an easy way to implement what you describe. The "move" command:
has the restriction that it can only "teleport" an atom not farther than one patch/domain away. Unless you're running at much fewer processors than optimal, each membrane would be more than one patch thick for sure.

Otherwise, simulations with exchanges of topology are fairly routine in NAMD and controlled by Tcl scripting commands (as done for example in constant pH simulations). One could have an array of topologies with individual pairs of ions swapped, and it'd be just the number of combinations to make this step tedious (need a folder full of hundreds of PSF files).


On Mon, Aug 5, 2019 at 7:12 AM Villalain Boullon, Jose <<>> wrote:
Dear all,

I have seen several papers using computational electrophysiology using Gromacs,>.

Do you know if such a methodology exist for NAMD ?. Or any reference in order to try to do it using NAMD?.

Thanks a lot !!.


Universidad "Miguel Hernández"
Alicante, Spain

Giacomo Fiorin
Associate Professor of Research, Temple University, Philadelphia, PA
Research collaborator, National Institutes of Health, Bethesda, MD

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2019 - 23:20:54 CST