Giacomo Janson

aSAM, a latent diffusion model trained on molecular dynamics, generates heavy-atom protein ensembles; its temperature-conditioned variant aSAMt reproduces mdCATH thermal unfolding and extrapolates beyond training temperatures.

This paper introduces GSnet, a graph neural network for predicting protein molecular properties, and shows it outperforms existing methods for solvent-accessible surface area prediction.

This paper introduces idpSAM, a deep generative model for transferable modeling of intrinsically disordered protein conformations, showing high performance on unseen sequences.

This paper improves MODELLER's 3D protein modeling by modifying its estimation of structural divergence and incorporating statistical potentials, enhancing prediction accuracy.