Sean Burgess

My research interests cover three broad topics:

1. Mechanistic chromatography modeling. I strive to understand how physics-based models can be applied to ever-expanding resin and molecule modalities. Model applications include both early and late-stage process development. I am also interested how Bayesian techniques, such as Markov Chain Monte Carlo, can help deepen understanding of calibration uncertainty and process robustness.
2. Machine learning-based interrogation of protein binding characteristics. Molecular process development can be an extremely expensive process, due, in part, to working on molecules that do not make it to the clinic, whether because they do not meet their endpoint, or are too difficult to develop. I am interested in using machine learning to reduce the experimental burden on scientists, and, more importantly, increase the likelihood that molecule development is rapid and successfully. I am interested in aspects of both automation and multiplexed analytics for high-throughput data generation and challenges related to predictive modeling.
3. Bayesian optimization and its applications to chromatographic peak deconvolution and characterization. In order to interpret spectral/chromatographic signals common in analytical chemistry, two common problems need to be solved. First, often signals from multiple protein species are hidden/convoluted in a chromatogram so the number of species and their relative abundance cannot be easily determined. Second, chromatogram changes and outlier detection are often critical to identify process robustness. I am interested in the development of methods and algorithms for removing sources of noise in the signal, and signal deconvolution.

• Rutgers University, New Brunswick, NJ, PhD in Chemical Engineering – 2019
• Rutgers University, New Brunswick, NJ, M.S. in Chemical Engineering – 2016
• Stony Brook University, Stony Brook, NY, B.S. in Chemical Engineering – 2011