Analysis of the experimental binding thermodynamics for approximately 100 protein-ligand complexes provides important insights into the factors governing ligand affinity and efficiency. The commonly accepted correlation between enthalpy and -TΔS is clearly observed for this relatively diverse data set. It is also clear that affinity (i.e., ΔG) is not generally correlated to either enthalpy or -TΔS. This is a worrisome trend since the vast majority of computational structure-based design is carried out using interaction energies for one, or at most a few, ligand poses. As such, these energies are most closely comparable to enthalpies not free energies. Closer inspection of the data shows that in a few cases the enthalpy (or -TΔS) is correlated with free energy. It is tempting to speculate that this could be an important consideration as to why some targets are readily amenable to modeling and others are not. Additionally, analysis of the enthalpy and -TΔS efficiencies shows that the trends observed for ligand efficiencies with respect to molecular size are primarily a consequence of enthalpic, not entropic, effects.
Keywords: Ligand efficiency; drug design; enthalpy entropy compensation; free energy; ligand binding; structure-based design.