College of Science & Engineering
Twin Cities
These researchers engineer high affinity binding proteins, enzymes, and antimicrobial proteins using laboratory directed evolution. Hundreds of millions of protein mutants are produced and tested for functionality (stability and affinity to various targets of interest, catalytic efficiency, and antimicrobial activity) on a weekly basis. Yet this seemingly high throughput does not approach the 10N (N=20-1,000) possible mutants. The group uses computational estimation of protein stability and protein-protein affinity to guide their selection of which mutants to test in the laboratory.
The researchers use several software packages including Rosetta, AlphaFold, and custom codes in Python and Linux. Multiple experimental questions will be asked:
- How do protein topology, paratope, and sequence impact developability and evolvability?
- What protein sequence motifs are universally effective?
- Are protein developability and evolvability predictable by natural homologs and structural data?
- What algorithms are most efficient at modeling sequence-function relationships?