Dr. Justin Siegel begins this episode by explaining what enzymes are, how they have evolved, and why Dr. Siegel is motivated to try to engineer enzymes to perform functions tailored to help humanity instead of to perform functions based on how they evolved in nature. He explains the primary goal of the work discussed and relating enzyme sequence to function. Dr. Siegel also explains how his work was the first of its kind by scaling up enzyme design to hundreds of mutants instead of dozens.  We then dig into the details of Dr. Siegel’s work. We learn details of his study such as why his team chose to study the particular enzyme that was used to create a massive set of enzyme mutants. We hear the previous difficulty of doing a study like this on only one enzyme and what enabled this increase in the scale of enzyme design. We also hear about how the use of cloud labs was introduced into the project and why.  Next, we hear all about the cloud lab aspect of this project. Dr. Siegel explains which parts of the enzyme mutant creation process were most challenging and benefited most to be moved to cloud labs.  Finally, we learn about how machine learning was then applied to the large set of generated enzyme mutants. Dr. Siegel explains how the generated data allowed his team to test previous hypotheses about mutant enzymes and to start trying to predict the functions of enzymes from sequence. Dr. Siegel also comments on a finding of the paper that for highly conserved residues, if you change them, you lose the function.  Learn more about Dr. Siegel’s work by reading the corresponding publication which you can find here: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0147596