Protein Engineering by Efficient Sequence Space Exploration Through Combination of Directed Evolution and Computational Design Methodologies

Directed evolution has matured in academia and industry as a versatile algorithm to redesign enzymes to match demands in biotechnological applications (as documented by the Nobel Prize in chemistry in 2018). Based on the obtained knowledge, computational methods (e.g. FRESCO, FoldX, CNA, PROSS, ProSAR) emerged to be predictive methods to especially improve properties that could be localized within a protein (e.g. thermostability, selectivity, catalytic efficiency, and activity). The main limitation to efficiently explore and benefit from nature's potential in generating better enzymes is the size of the protein sequence space; experimentalists have to admit that they will never be able to experimentally sample through the whole sequence space. A combination of experimental and computational methods proved to be time efficient in redesigning enzymes to meet the application demands (e.g. in chemical and pharmaceuticals synthesis). In this chapter, we highlighted protein engineering strategies that combine directed evolution and computational analysis to efficiently reengineer enzymes and that partly contribute to a molecular understanding of structure function relationship, which can be transferred from enzymes to another. In this respect, an emphasis will be given to the “KnowVolution” (knowledge gaining directed evolution) strategy, which is generally applicable, minimizes experimental efforts, generates a molecular understanding on each positions/amino acid substitution, and was successfully applied to a broad range of and enzymes and properties.