LiveCoMS Roundtable: Discussing Markov state models after two decades of development v1.0

Markov state models (MSMs) play a central role in the analysis of molecular dynamics (MD) simulations for biomolecules, and MSM methodologies have been under continual development since their introduction to the field in the early 2000s. MSMs take MD trajectory data as input and produce a more compact, i.e., coarse-grained, and interpretable representation of the system in terms of conformational states and quantitative observables such as the equilibrium populations of states and the timescales of interconverting between states. Numerous strategies and considerable mathematical formalism have been developed to improve coarse-graining in both time and space to enable the most reliable representation of the underlying biomolecular dynamics. Nevertheless, there remains considerable discussion in the literature regarding best practices and MSM limitations, as well as proposals to abandon or extend the first-order Markovian framework in common use. This Roundtable article gathers together diverse opinions from $> 10$ authors active in MSM development, and by design, each author or author group offers a separate contribution. No consensus view is provided. Instead, the article is meant to be a guide for readers wishing to be exposed to a wide set of expert perspectives.