Yasuhiro Matsunaga

This paper fine-tunes a protein language model on nanobody thermostability data and uses sparse autoencoders to uncover biophysical rules governing stability, identifying both known and novel stabilizing features.

The paper predicts nanobody melting temperatures using transfer learning with computed stability data, finding that free-energy perturbation labels and sequence diversity improve predictions.

AFM-Fold reconstructs 3D protein conformations from AFM images using a generative AI framework, outperforming rigid-body fitting and capturing conformational dynamics.

The paper introduces a framework to describe viral capsid subunit shapes using spherical tiling theory, revealing distinct interface groups and potential assembly pathways in T = 1 capsids.