On Mon, Jan 23, 2017, Andreas Tosstorff wrote:
>
> I saw different approaches on how to calculate free binding energies
> from PMFs in the literature:
>
> Some took the difference between the PMF curve's energy minimum and the
> value at the maximum distance.
This is certainly a pretty drastic approximation, since the width of the
"well" containing the bound ligand can be just as important as the energy at
the bottom of the well.
If you know the PMF accurately (a big "if") then you can compute the
probability of the system being at any particular value of the progress
variable. Next: figure out a dividing point, so that you are willing
to call configurations that are on one side of the dividing point "bound" and
on the other side "unbound". Finally, compute an equilibrium constant by
dividing the fraction bound by the fraction unbound, and convert this
equilibrium constant to a standard free energy in the usual way.
[Caveat: above is a bit oversimplified for ligand binding, since PMF's are
generally computed with a *single* ligand and a single receptor; you have to
make the usual standard state corrections to account for the concentration
dependence.]
Other methods use different (often more efficient) ways to estimate the work
required to dissociate the ligand from the receptor. But the procedure
outlined in the previous paragraph is the only one (I think) that fits the
description of computing the binding free energy *from* the PMF.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jan 24 2017 - 06:00:03 PST
