My curiosity in neuroscience and AI started with a simple question: how can we understand the brain and accelerate discoveries in neurodegenerative diseases? Over the last four years, I’ve explored this question through personal projects, internships, and AI modeling, building a path from early experiments to real-world research. Below is a description of my two most important neuroscience research projects.
Amyloid-Beta Inhibitor Generation via AI Learning (ABIGAIL):
ABIGAIL is a graph neural network. This neural network generates novel small molecules that inhibit amyloid-beta, which is a protein that can speed up Alzheimer’s development. What I did uniquely with ABIGAIL is I introduced a 3rd dimension in the latent space (where the model finds these molecules), thus making it more accurate. I hope that ABIGAIL can be used to find potential drug therapies for Alzheimer’s.
Deriving a Novel Index For a More Objective
Prediction of Parkinson’s
During my time in the RISE summer computational neurobiology, I spent much of my time on this project: deriving an equation that solved for an index that could predict Parkinson’s better than the current standard index, UPDRS. My team and I ended up creating two seperate index formulas: one for clinical use, and one as an at-home test.
Behind the Brain 