G protein-coupled receptors (GPCRs) transform signals from a wide range of stimuli such as odorants, food, hormones, neurotransmitters, light, pain, digestion, and cardiovascular regulation into cellular responses. GPCRs have served as important targets for treating many human diseases, including opioid addiction, depression, obesity, heart failure, cancer, and diabetes. Robust and reliable methods for the prediction of binding free energies are of high value to the GPCR drug discovery process. The very recent arrival of AMBER GPU accelerated Thermodynamic Integration (TI) facilitates very fast, highly accurate binding affinity predictions for small molecules and peptides. An automated free energy calculation platform is being built at Alkermes to set up high performance GPU TI calculations using AMBER, FESetup, LOMAP, and workflow tools such as Pipeline Pilot and KNIME. Aided by the significant speedup from using GPUs, and enabled by a growing number of high-resolution GPCR crystal structures, rigorous free energy calculations can be applied in rational drug design. Here we evaluated the performance and accuracy of a binding free energy calculation workflow on multiple sets of GPCR ligands for mu, kappa and delta opioid receptors. We believe the robustness and speed of the workflow will help drive the discovery of new medicines and accelerate the pace of GPCR drug discovery.

Rapid and accurate binding affinity prediction for GPCR ligands through AMBER GPU accelerated thermodynamic integration

256th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.