Machine learning-assisted directed protein evolution with combinatorial libraries
Machine learning-assisted directed protein evolution with combinatorial libraries is a scholarly work by Russell D Lewis and Frances Arnold, published in 2019 in ''Proceedings of the National Academy of Sciences of the United States of America''. The main subjects of the publication include machine learning, metabolic engineering, computer science, microfluidics, workflow, directed evolution, chemical space, computational biology, sequence space, immobilized enzyme, biological sequence, protein engineering, in silico, protein sequencing, and artificial intelligence. The authors validated this approach on a large published empirical fitness landscape for human GB1 binding protein, demonstrating that machine learning-guided directed evolution finds variants with higher fitness than those found by other directed evolution approaches.