PhD Researcher – Generative AI & Molecular Dynamics
As a PhD researcher, I built, trained, and validated machine learning models on molecular dynamics datasets to identify protein conformational states. My work involved generating and curating labels representing metastable binding-competent conformations and binding modes relevant for drug discovery. These tasks contributed to the development of AI systems capable of distinguishing complex biophysical behaviors in proteins and RNA. • Developed and labeled training datasets from molecular dynamics (MD) simulations for use in VAE, VDE, and Random Forest models. • Curated and classified fuzzy versus specific protein binding modes using machine learning. • Compared latent representations to annotate disorder–order transitions in IDPs. • Simulated and modeled protein–RNA stability with labeled kinetic and structural features.