Abstract: Engineered microbes are masters of matter-energy-information conversion. With their help, we can overcome several 21st century crises, for example, by reversing climate change, identifying harmful chemicals, and manufacturing high-value products from low-cost renewable feedstocks. However, engineering organisms to achieve a high-performance function still requires too much trial-and-error, due to the many interacting components (sensors, enzymes, regulators) working together inside a cell. New efforts are needed to transition Synthetic Biology towards becoming a mature engineering discipline where genetic systems are reliably designed with high-level specifications.The Salis lab combines computational and experimental approaches to create a "predictive model stack" and integrated design platform for engineering microbes. We apply our design platform to engineer sensors, genetic circuits, and metabolic pathways to reprogram organisms with complex functions. For example, we recently engineered soil bacteria to sense the TNT explosive and maintain stable detection for over 21 days in a contaminated soil environment filled with natural microbes. We develop and validate predictive models using massively parallel design-build-test workflows that enable us to efficiently collect thousands of experimental measurements. We combine physiochemical modeling (statistical thermodynamics & kinetics) with machine learning to formulate and parameterize sequence-to-function models, which are validated on large datasets. With this approach, we have developed models for controlling the rates of transcriptional initiation, translation initiation, translational coupling, mRNA decay, riboswitch function, and more. We then apply multi-objective optimization with newly developed path generation algorithms to automatically design genetic systems that maximally satisfy several metrics and rules, encompassing function, genetic stability, and buildability. We present several genetic systems engineering examples to illustrate our automated design approach.
Bio: Prof. Howard Salis is a Professor in the Department of Chemical & Environmental Engineering at the University of California Riverside. He obtained his Ph.D. degree in Chemical Engineering from the University of Minnesota and completed a postdoctoral fellowship at UC San Francisco. His research lab develops predictive models and design algorithms to rationally engineer organisms for industrial and medical biotech applications, formulated using thermodynamics & kinetics and tested using massively parallel experiments. He received the DARPA Young Faculty Award in 2010, the NSF Career Award in 2013, and founded De Novo DNA, a spin-off company that commercializes design methods for genetic systems.
Abstract: Engineered microbes are masters of matter-energy-information conversion. With their help, we can overcome several 21st century crises, for example, by reversing climate change, identifying harmful chemicals, and manufacturing high-value products from low-cost renewable feedstocks. However, engineering organisms to achieve a high-performance function still requires too much trial-and-error, due to the many interacting components (sensors, enzymes, regulators) working together inside a cell. New efforts are needed to transition Synthetic Biology towards becoming a mature engineering discipline where genetic systems are reliably designed with high-level specifications.The Salis lab combines computational and experimental approaches to create a "predictive model stack" and integrated design platform for engineering microbes. We apply our design platform to engineer sensors, genetic circuits, and metabolic pathways to reprogram organisms with complex functions. For example, we recently engineered soil bacteria to sense the TNT explosive and maintain stable detection for over 21 days in a contaminated soil environment filled with natural microbes. We develop and validate predictive models using massively parallel design-build-test workflows that enable us to efficiently collect thousands of experimental measurements. We combine physiochemical modeling (statistical thermodynamics & kinetics) with machine learning to formulate and parameterize sequence-to-function models, which are validated on large datasets. With this approach, we have developed models for controlling the rates of transcriptional initiation, translation initiation, translational coupling, mRNA decay, riboswitch function, and more. We then apply multi-objective optimization with newly developed path generation algorithms to automatically design genetic systems that maximally satisfy several metrics and rules, encompassing function, genetic stability, and buildability. We present several genetic systems engineering examples to illustrate our automated design approach.
Bio: Prof. Howard Salis is a Professor in the Department of Chemical & Environmental Engineering at the University of California Riverside. He obtained his Ph.D. degree in Chemical Engineering from the University of Minnesota and completed a postdoctoral fellowship at UC San Francisco. His research lab develops predictive models and design algorithms to rationally engineer organisms for industrial and medical biotech applications, formulated using thermodynamics & kinetics and tested using massively parallel experiments. He received the DARPA Young Faculty Award in 2010, the NSF Career Award in 2013, and founded De Novo DNA, a spin-off company that commercializes design methods for genetic systems.