Toward Automated Physics-Based Absolute Drug Residence Time Predictions

The residence time (τ) of a drug bound to a receptor target is increasingly recognized as a key property to control during ligand-optimization campaigns and provides a useful dimension for modulating compound profiles in addition to binding affinity. Here we propose a new scheme to calculate absolute residence times by using two enhanced sampling approaches, consisting of an exploration phase followed by an exploitation phase that estimates the residence time: Random Acceleration Molecular Dynamics (RAMD) [Lüdemann S.K. <i>et al.</i>, 2000] to harvest plausible egress pathways, and then Infrequent Metadynamics (iMetaD)² to estimate residence time. This protocol caters to small molecule drug discovery programs, where a key aspect is the compromise between accuracy, throughput, and ease of use. We benchmark this approach by computing residence times for five diverse drug targets, each with its own congeneric series of compounds, for a total of 29 drug-target complexes. When comparing computed residence times with experimental values, good accuracy (RMSE of 1.22 and <i>R</i>² of 0.80 in log₁₀(τ)) is achieved without manually tuning the enhanced sampling parameters.