From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Tue Sep 13 2005 - 11:31:22 CDT
Hi Ana,
this value is in kcal/mol, but if you want to correlate to experimental
values you need to calculate something to compare it to; just taking the
sum of the energy interactions tells you very little on its own unless
all you're doing is comparing small differences in binding mode. For
example, if you have two proteins P1 and P2 interacting with each other,
you can calculate their enthalpy of binding as
?H(binding) = (E(P1,P2) + E(P1P2,solvent) + E(P1) + E(P2)) -
(E(P1,solvent) + E(P2,solvent) + E(P1) + E(P2))
With E(i,j) defined as the interaction energy between i and j, and E(i)
is defined as the internal energy of i.
Where the first sum term is calculated on the equilibrated bound
structure, and the last sum term is calculated for the equilibrated
structures of the two separated proteins (often you can make the
approximation to ignore, for example, the internal energies, if the
protein conformations don't change much upon binding).
Even the enthalpy of binding probably doesn't correspond to the
experimental properties you're looking for because it doesn't include
entropic terms. If you're comparing to binding constants, for example,
you'd need to calculate the free energy of binding instead, which isn't
easy for two large structures (you could, for example, simulate the
pulling of the two subunits apart from each other several times and try
to apply Jarzinsky's equality). The values for enthalpy of binding may
match up with binding constants if there is little entropic change upon
binding, and should certainly provide a more resonable estimate of the
energy change upon binding.
Peter
Anna Modzelewska wrote:
> Hi,
> I calculated the pair interactions energy between two proteins and I
> got values about -2500. Is it in kcal/mole?
> If yes, why these energies are so big? How to correlate them with
> experimental values?
>
> Thank you
> Anna
>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:39:55 CST