Generative Biology Reading Group - Community Only
Starkly Speaking is a weekly reading group hosted by Hannes Stärk (MIT) where paper authors present and discuss their own work, every Monday at 9am PT / 12pm ET / 6pm CET. The series sits at the intersection of machine learning and the molecular sciences, with recurring themes including generative modeling for biology and chemistry (diffusion models, flow matching, classifier-free guidance), protein structure and conformational ensemble generation, molecular dynamics surrogates, neural network potentials for atomistic systems, and single-cell / perturbation modeling. Recent sessions have covered topics like universal models for atoms (UMA), continuous-denoising language models, cellular perturbation prediction, and protein generation methods such as ProtComposer and AlphaFlow. This session: MarS-FM — Generative Modeling of Molecular Dynamics via Markov State Models (Kapuśniak, Gabellini, Bronstein, Tossou, Di Giovanni — arXiv:2509.24779) Molecular dynamics is a powerful but expensive computational microscope for studying protein function, since fine-grained integration over long biological timescales is costly. Generative surrogates help, but most learn a fixed-lag transition density, so training is dominated by frequent, uninformative transitions. This paper introduces a new class of generative models — MSM Emulators — that instead learn to sample transitions between the discrete states defined by an underlying Markov State Model. The authors instantiate it as Markov Space Flow Matching (MarS-FM), which achieves more than two orders of magnitude speedup over implicit- and explicit-solvent MD. They evaluate how well it reproduces MD statistics (RMSD, radius of gyration, secondary structure) across diverse protein domains up to 500 residues — including unfolding events — under strict train/test sequence-dissimilarity to test generalization, where MarS-FM outperforms existing methods, often substantially.
This event is hosted at the Frontier Tower:
We are transforming a 16-floor tower in San Francisco into a self-governed vertical village—a hub for frontier technologies and creative arts. Tier-one labs presenting AI, Ethereum, biotech, neuroscience, longevity, robotics, makerspace, human flourishing, and arts & music. These floors will house innovators and creators pushing the boundaries of human potential in a post-AI-singularity world.
Apply here for founding citizenship: https://frontiertower.io/apply
Why should I become a citizen?
Be part of creating the first self-governed vertical village
Connect with the most creative people in the city
Get access to all floors, free event space & movement floor
Website: https://frontiertower.io/
Need more reading? Visit https://frontiertower.notion.site/