Predicting Biomolecular Binding Kinetics: A Review

Abstract

Biomolecular binding kinetics including the association (k on ) and dissociation (k of f ) rates are critical parameters for therapeutic design of smallmolecule drugs, peptides, and antibodies. Notably, the drug molecule residence time or dissociation rate has been shown to correlate with their efficacies better than binding affinities. A wide range of modeling approaches including quantitative structure-kinetic relationship models, Molecular Dynamics simulations, enhanced sampling, and Machine Learning has been developed to explore biomolecular binding and dissociation mechanisms and predict binding kinetic rates. Here, we review recent advances in computational modeling of biomolecular binding kinetics, with an outlook for future improvements.

Characterization of binding kinetics and intracellular signaling of new psychoactive substances targeting cannabinoid receptor using transition-based reweighting method

“…These observations agree with the previously reported computational research, where experimentally comparable estimation of k off rates were shown to be more challenging compared to k on . 63…”

A multiscale predictive digital twin for neurocardiac modulation

“…Since ligand binding to the adrenergic receptor and other G protein‐coupled receptors does not follow a linear path (Provasi et al., 2009; Schneider et al., 2015) as a drug binding to the hERG channel pore (DeMarco et al., 2021; Yang et al., 2020), we used a different enhanced sampling molecular dynamics simulation approach, well‐tempered metadynamics, and estimated dissociation rate using Kramer rate theory formalism based on computed free energy profile and diffusion coefficient at the free energy barrier computed from a separate restrained molecular dynamics simulation. Calculation of diffusion coefficient profiles across an entire reaction coordinate as was done in previous studies (Berezhkovskii et al., 2011; DeMarco et al., 2018; DeMarco et al., 2021; Setny et al., 2013; Vorobyov et al., 2014; Yang et al., 2020; Zhu & Hummer, 2010; Zhu & Hummer, 2012) can help refine computed ligand–receptor ‘on’ and ‘off ’ rates and will be explored in our subsequent studies along with other rate computation methods (Meral et al., 2018; Palacio‐Rodriguez et al., 2022; Pang & Zhou, 2017; Wang et al., 2023). This methodological shortcoming may contribute to a substantial difference between our MD estimated and substantially slower experimental ‘on’ and ‘off ’ rates from a recent study (Xu et al., 2021).…”

“…This methodological shortcoming may contribute to a substantial difference between our MD estimated and substantially slower experimental ‘on’ and ‘off ’ rates from a recent study (Xu et al., 2021). Achieving agreement between computed and experimental protein‐ligand rates has proven to be a challenging task with results varying by several orders of magnitudes and discrepancies attributed to issues with enhanced sampling methodologies, choice of reaction coordinate (s), force field accuracy, presence of hidden barriers and other MD simulation related factors (Wang et al., 2023) as well as those influencing accuracy of experimental rate measurements using, e.g. competitive binding assay (Georgi et al., 2019; Hoare, 2021; Sykes et al., 2019).…”

Kinetics from Metadynamics: Principles, Applications, and Outlook

“…In the field of biology, most applications of enhanced sampling approaches for kinetics calculation have focused on the binding/unbinding kinetics of protein–ligand complexes. − This can be attributed to the increased realization that ligand residence time is a key predictor of drug efficacy. − Therefore, efforts have been dedicated to the development and application of computational methods to predict the drug–receptor unbinding rate constant ( k off ). In Table , we summarize the protocol and results of such studies involving metadynamics-based approaches.…”