(568g) Towards a Predictive Synthetic Biology Enabled By Machine Learning and Automation

Biology has been transformed in the second half of the 20th century from a descriptive to a design science. We can engineer cells faster than ever, enabled by exponentially growing DNA synthesis and revolutionarily effective tools like CRISPR-enabled gene editing. However, while we can make the DNA changes we intend, the end result on cell behavior is usually unpredictable. In this talk, I will explain our efforts to create predictive algorithms that take -omics data and produce actionable items for bioengineering biofuel producing cells. I will show how machine learning and mechanistic models, enabled by automation capabilities such as microfluidics, can produce predictions accurate enough to drive synthetic biology efforts. While a fully predictive biology is not within easy reach, we believe the bases for making it a reality in the next 50 years are set.