The research group was funded in 2026, thanks to the University Excellence Fund. The main objective of the research group is to develop a next-generation framework that integrates Artificial Intelligence (AI) with Molecular Dynamics (MD). Our goal is to gain a deeper understanding of the complex machinery behind biological systems, specifically focusing on multi-state proteins and intrinsically disordered regions. By combining generative models with energy learning based on the Boltzmann distribution, our method accelerates conformational sampling while strictly maintaining physical accuracy. This hybrid approach allows us to efficiently map the high-dimensional energy landscapes of proteins that have previously been difficult to simulate.

We are applying this new technology to investigate redox-sensitive disordered regions and the large-scale conformational transitions of viral fusion proteins. Through these studies, we hope to uncover critical insights into the mechanics of protein regulation and the physical processes underlying viral infections. Ultimately, the project aims to democratize these tools by developing a fully open-source AI-MD platform for the broader scientific community.