Does a Machine-Learned Potential Perform Better Than an Optimally Tuned Traditional Force Field? A Case Study on Fluorohydrins.

We present a comparative study that evaluates the performance of a machine learning potential (ANI-2x), a conventional force field (GAFF), and an optimally tuned GAFF-like force field in the modeling of a set of 10 γ-fluorohydrins that exhibit a complex interplay between intra- and intermolecular interactions in determining conformer stability. To benchmark the performance of each molecular model, we evaluated their energetic, geometric, and sampling accuracies relative to quantum-mechanical data. This benchmark involved conformational analysis both in the gas phase and chloroform solution.

Discovery of a Potent and Orally Bioavailable Zwitterionic Series of Selective Estrogen Receptor Degrader-Antagonists.

Herein, we report the optimization of a meta-substituted series of selective estrogen receptor degrader (SERD) antagonists for the treatment of ER+ breast cancer. Structure-based design together with the use of modeling and NMR to favor the bioactive conformation led to a highly potent series of basic SERDs with promising physicochemical properties. Issues with hERG activity resulted in a strategy of zwitterion formation and ultimately in the identification of .