Autodock Vina gives the estimated binding energy of the docked ligand at the active site of the receptor.
In my case, binding glycine to different target variants (more than 20 slightly different variants), I was not able to obtain binding energies lower than -4.1 kcal/mol. I suspect that this value is the lowest theoric binding energy that could be obtained with glycine independently of the target protein; i.e. it is not possible to obtain lower binding energies even with further optimized target variants or, in general, any protein. This is due to the fact that the number of non-covalent interactions of glycine are finite.
Can someone confirm this hypothesis. It there a "simple" method to estimate the theoretic total binding energy of a small compound so I could confirm the value I obtained?
I don't believe this is the case. When you consider a P + L PL equilibrium, the free energy variation of the system depends on both the enthalpy and entropy variation of both the protein and the ligand. Thus, although one may anticipate a limiting entropic variation for the desolvation and reduction of degrees of freedom of the ligand upon binding, this contribution will differ in different receptors. Also, the enthalpic contribution to the binding is affected by long range effects, not simply the number and geometry of direct hydrogen bonds or van der Waals' contacts.
You may try using molecular dynamics and thermodynamic cycles of (some of) your docked protein-ligand complexes as an independent means to calculate binding energies to your protein in water. This way, you will take into account the ligand and protein flexibility, use a different force field and electrostatic calculations, and have an independent estimate of the dissociation constants.
- Badges
- Science topic
- Similar topics
- Engineering
- Petroleum Engineering
- Oil