Protein Folding is the "molecular origami" that is required for nearly all of our cells' molecular machines (proteins) to function. Although the recipe for building our body's proteins is written in our DNA, the process of going from a sequence of parts to a functional protein is incredibly complex. Despite more than a half-century of study, the "protein folding problem" remains unsolved. Although recent breakthroughs in AI have allowed accurate prediction of what a protein will look once it is folded, we still cannot accurately predict how long it will take to fold and when it will fail (let alone why). This failure lies at the heart of many terrible diseases including Alzheimer's Disease and some forms of cancer. We believe that the answer to this question lies in the study of this problem on multiple length and timescales, many of which are only accessible to physics-informed molecular simulation with the aid of powerful computers (or networks of computers such as in the folding@home project).

Computational Drug Discovery is the process of using computer simulations to identify or design useful pharmaceutical compounds. Our goal is to develop a deep understanding of the physics behind protein-ligand interactions and use this knowledge to predict the activity of new drug compounds. In our group we employ physics-based methods such as docking, free-energy perturbation, and molecular dynamics along with deep learning and topological data analysis to accomplish this task.

We are committed to the mission of lifelong learning, we enjoy spending time studying other topics with less obvious and immediate applications to computational chemistry, biophysics, and materials science. Of particular interest to our group are artificial intelligence and quantum computing. While these areas have significant practical importance, they also offer a glimpse into the deep foundational scientific questions that have captivated humanity since antiquity.