seekrflow: Towards an End-to-End Automated Simulation Pipeline with Machine-Learned Force Fields for Accelerated Drug-Target Kinetic and Thermodynamic Predictions.

Accurate prediction of drug-target binding and unbinding kinetics and thermodynamics is essential for guiding drug discovery and lead optimization. However, traditional atomistic simulations are often computationally expensive to capture rare events that govern ligand (un)binding. Several enhanced sampling methods exist to overcome these limitations, but they require extensive manual intervention and introduce variability and artifacts in free energy and kinetic estimates that limit high-throughput scalability. The present work introduces seekrflow, an automated multiscale milestoning simulation pipeline that streamlines the entire workflow from a single receptor-ligand input structure to kinetic and thermodynamic predictions in a series of steps. This integrated approach minimizes manual intervention, reduces computational overhead, and enhances the reproducibility and accuracy of kinetic and thermodynamic predictions. The accuracy and efficiency of the pipeline are demonstrated across multiple receptor-ligand complexes, including inhibitors of heat shock protein 90, threonine-tyrosine kinase, and trypsin, with predicted kinetic and thermodynamic parameters that closely match experimental estimates. seekrflow establishes a new benchmark for automated and high-throughput physics-based predictions of the kinetics and thermodynamics.