Assessing the Accuracy and Performance of Implicit Solvent Models for Drug Molecules: Conformational Ensemble Approaches

Abstract

The accuracy and performance of implicit solvent methods for solvation free energy calculations were assessed on a set of 20 neutral drug molecules. Molecular dynamics (MD) provided ensembles of conformations in water and water-saturated octanol. The solvation free energies were calculated by popular implicit solvent models based on quantum mechanical (QM) electronic densities (COSMO-RS, MST, SMD) as well as on molecular mechanical (MM) point-charge models (GB, PB). The performance of the implicit models was tested by a comparison with experimental water-octanol transfer free energies (ΔG(ow)) by using single- and multiconformation approaches. MD simulations revealed difficulties in a priori estimation of the flexibility features of the solutes from simple structural descriptors, such as the number of rotatable bonds. An increasing accuracy of the calculated ΔG(ow) was observed in the following order: GB1 ~ PB < GB7 ≪ MST < SMD ~ COSMO-RS with a clear distinction identified between MM- and QM-based models, although for the set excluding three largest molecules, the differences among COSMO-RS, MST, and SMD were negligible. It was shown that the single-conformation approach applied to crystal geometries provides a rather accurate estimate of ΔG(ow) for rigid molecules yet fails completely for the flexible ones. The multiconformation approaches improved the performance, but only when the deformation contribution was ignored. It was revealed that for large-scale calculations on small molecules a recent GB model, GB7, provided a reasonable accuracy/speed ratio. In conclusion, the study contributes to the understanding of solvation free energy calculations for physical and medicinal chemistry applications.

The Semiempirical Quantum Mechanical Scoring Function for In Silico Drug Design

“…[84] ) In our score, conductor-like screening model (COSMO) [85] and SMD [86] are used, because these methods describe the hydration free energy most accurately for drug molecules. [87] However, the SMD method is so computationally demanding that it is used only for the ligand, giving rise to the correction of ligand solvation/desolvation (DG low solv (L)ÀDG high solv (L)). It is the difference between the solvation free energy calculated at high (SMD) and low (COSMO) levels of theory.…”

“…The current best practice is to use the COSMO model, [87] which is implemented in the MOPAC2012 code. It can be used in conjunction with SQM methods and is applicable even for large complexes.…”

“…To speed up the calculations, we have also proposed a SQM/MM procedure [81] in which protein surroundings up to 8 from the ligand were treated by SQM, whereas the rest of the protein was described by MM and surrounded by water on a less accurate and fast GB implicit solvent level. [87] In such a setup, however, caution must be taken regarding amino acid ion pairs, because unphysical proton transfers can occur due to incomplete polarisation of the SQM part from the MM part as well as due to the underestimation of the proton-transfer barrier in SQM. [91,44] …”

“…[69] It should be noted that the series of HIV PR inhibitors consisted of large and flexible molecules, either neutral or charged, which present a challenge for computations. [87,92] The DG 0w conf (L) term was calculated by two simple and fast approaches, that is, by simple SQM minimisation and SQM-based MD/quenching. The differences between the calculated DG 0w conf (L) were, however, small for most ligands.…”

“…The electronic gas-phase interaction energy (PM6-D3H4X) [44,45] and the implicit solvation method (SMD, COSMO or COSMO-RS) [87] have probably reached the limit for neutral molecules, but their performance for charged molecules remains to be fully assessed. The ligand-deformation (strain) prediction depends not only on the accuracy of the gas-phase and solvation-energy method but also on the ability to find a (set of) energetically low-lying structures.…”