(117g) Data Driven Biomolecular Engineering Using High Throughput Measurements of Protein Function

Massively parallel, high throughput measurements of protein function have revolutionized protein science and engineering. Such measurements enable machine learning and other data-driven approaches for the efficient navigation of functional sequence space. In this talk I will describe the technology my lab has built for determining quantitative sequence-function relationships in enzymes for tens of thousands of variants using growth-based selections and screens. Then, I will show data-driven insights for improvement of enzymes involved in biomass catabolic pathways, secondary metabolic pathways, and fungal bioluminescence. I’ll end the talk by describing a complementary bottoms-up strategy to regulate enzymes using small molecules, with the engineering enabled by high throughput screening.