A novel method for calculating relative free energy of similar molecules in two environments

Abstract: Calculating relative free energies is a topic of substantial interest and has many applications including solvation and binding free energies, which are used in computational drug discovery. However, there remain the challenges of accuracy, simple implementation, robustness and efficiency, which prevent the calculations from being automated and limit their use in computational drug discovery. Here we present an exact and complete decoupling analysis in which the partition functions of the compared systems decompose into the partition functions of the common and different subsystems. This decoupling analysis is applicable to submolecules with coupled degrees of freedom and to any potential function, enabling to remove less terms in the transformation. Then we show mathematically, in the context of partition function decoupling, that the two compared systems can be simulated separately, eliminating the need to design a composite system. We demonstrate the decoupling analysis and the separate transformations in an MD calculation of relative free energy for a general force field which is in agreement with experiments. We present a unified soft core technique that will ensure the monotonicity of the numerically integrated function. Finally, we show that when the systems have rugged energy landscape they can be equilibrated without introducing another sampling dimension. The concepts presented in the article have implications in accuracy, efficiency, simplicity and robustness of free energy calculations.