Re: ABF and Kinetic Barriers

From: Jérôme Hénin (
Date: Mon Aug 12 2019 - 05:43:00 CDT

Hi Kelly,

If kinetics along the pathway is as you describe determined by the
relaxation rate of another degree of freedom, then Eyring rate theory is
inadequate to describe that process.

More generally: free energy profiles are purely thermodynamic descriptions.
They only provide kinetic information under strong assumptions on the
nature of the transition, and on the reduced coordinates.


On Mon, 12 Aug 2019 at 02:33, McGuire, Kelly <> wrote:

> How well does ABF handle kinetic barriers? I have on and off rate
> constants for a ligand binding to a protein from electrophysiology. The on
> rate constant is ~370 - 400 M(-1)s(-1). This is a relatively low rate
> constant, suggesting that the ligand does not pass the entry barrier very
> frequently. The off rate constant however is very small (2x10^-5 s(-1)),
> suggesting that when the ligand gets passed the entry barrier, it remains
> in the binding site for a long time. The Kd is therefore in the low
> micromolar range. The barrier to entry in this protein is believed to be a
> cluster of valines, 4 of them to be exact. The protein is a homotetramer.
> I ran an ABF with the protein and the ligand, and the binding site well is
> right in the free energy range we expect relative to bulk water, ~ -13
> kcal/mol. However, the energy barrier at the valines in only 2.5-3
> kcal/mol. Eyring rate theory predicts the barrier to be ~10-12 kcal/mol
> with the on rate constant obtained from electrophsyiology.
> I used 17 windows for my ABF, each window is 3 angstroms wide. I ran each
> ABF window for 100 nanoseconds. We are concerned that running one of the
> windows with the ligand near the entry barrier (valines) and the some of
> the other windows near the valines for too long might over sample the
> energy at that region and capture only the thermodynamic free energy and
> not the free energy including the kinetic barrier. It is unlikely that the
> ligand sits at the valines for 100 ns. It probably enters/collides for a
> short time then leaves and comes back or a new ligand collides with the
> barrier, but having an ABF window with the ligand right at the valines for
> 100 ns is probably too much sampling.
> So is it possible to over sample some regions of the ABF reaction
> coordinate and get a lower energy than expected because the valines were
> allowed to relax around the ligand, or does ABF still capture the
> thermodynamic and kinetic energy barrier effects correctly?
> *Kelly L. McGuire*
> *PhD Candidate*
> *Biophysics*
> *Department of Physiology and Developmental Biology*
> *Brigham Young University*
> *LSB 3050*
> *Provo, UT 84602*

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2020 - 23:17:11 CST