Data-Driven Molecular Engineering

Understanding how the sequence of amino acids in a protein determines its function is one of the grand challenges of molecular biology—one that has resisted solution for decades. Evozyne is pioneering a brand new approach to this problem that overcomes the limitations of existing techniques. We use state-of-the-art statistical and machine learning algorithms to build predictive models of sequence-function relationships for a family of proteins, then test the models by synthesizing, expressing, and characterizing thousands of never-before-seen sequences using our proprietary technologies. This experimental data is then used to refine the computational models and the process is repeated in a virtuous cycle of improvement until robust models of protein function are obtained. Our model-guided approach incorporates and explores sequence space more efficiently than alternative approaches such as directed evolution and more accurately than physics-based design.

Our Workflow

To apply this process generally and at scale, Evozyne brings together three proprietary technologies into a single streamlined workflow.

protein design
de novo
protein design
feedback loop

Contact us to learn more directly from our scientific team