Modular Machine Learning for Genetic Circuit Engineering

Abstract

Physics-based models have been extensively used for modular design of genetic circuits. Modules are first characterized in “isolation” and then composed together to predict the emergent behavior of a system. Nevertheless, designs often fail in practice because models do not accurately predict the systems’ output. This is due to unintended interactions among modules arising from sharing limited cellular resources, growth effects, and DNA supercoiling to name a few. This context-dependence can be incorporated to some extent in biophysical models used for design, but many unknowns still remain. What if we try to learn context-dependence through machine learning (ML) models that yet retain a modular architecture to capture modules’ physical properties? In this talk, I will introduce a modular ML approach to learn modules and their context-dependence from data in “isolation” and when modules are together in one combination only. I will show the ability of these models to extrapolate to different modules combinations, which makes the approach suitable for forward engineering. I will demonstrate the experimental performance of this approach for engineering bacterial multiplexed biosensors, in which quantitative output accuracy, as opposed to on-off, is required.

Bio

Dr.Domitilla Del Vecchio received her Ph. D. in Control and Dynamical Systems from the California Institute of Technology, Pasadena, and the Laurea degree in Electrical Engineering (Automation) from the University of Rome at Tor Vergata in 2005 and 1999, respectively. From 2006 to 2010, she was an Assistant Professor in the Department of Electrical Engineering and Computer Science and in the Center for Computational Medicine and Bioinformatics at the University of Michigan, Ann Arbor. In 2010, she joined the Department of Mechanical Engineering at the Massachusetts Institute of Technology (MIT), where she is currently the Grover M. Hermann Professor in Health Sciences and Technology and a Professor of Mechanical and Biological Engineering. She was awarded a 2024 Vannevar Bush Faculty Fellowship, she is a Fellow of the International Federation of Automatic Control (2022), an IEEE Fellow (2021), a recipient of the Newton Award for Transformative Ideas during the COVID-19 Pandemic (2020), the 2016 Bose Research Award (MIT), the Donald P. Eckman Award from the American Automatic Control Council (2010), the NSF Career Award (2007), the American Control Conference Best Student Paper Award (2004), and the Bank of Italy Fellowship (2000). Her research focuses on developing modeling and biological engineering techniques to understand and control the behavior of genetic circuits in bacterial and mammalian cells. Her lab is particularly interested in applications to biosensing, biomanufacturing, and regenerative medicine.