Directed evolution has proven to be a powerful tool in the engineering of proteins, and in particular enzymes.  It is often the case that dozens of genetic mutations are required to optimize an enzyme’s performance for a specific application.  With most mutations being deleterious, the incorporation of so many modifications require a guided approach to traverse sequence space to ensure that only the most promising mutations are included.  Machine learning techniques can play a key role in navigating through the multi-dimensional landscape of sequence-activity space.  The power of software will be highlighted in the development of successful programs that have led to commercially relevant enzymes in the in the food, pharmaceutical and biotherapeutic fields.

Use of machine learning to navigate the sequence-activity landscape during directed evolution campaigns

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.