Directed evolution (DE) creates diversity in subsequent rounds of mutagenesis in the quest of increased protein stability, substrate binding, and catalysis. Although this technique does not require any structural/mechanistic knowledge of the system, the frequency of improved mutations is usually low. For this reason, computational tools are increasingly used to focus the search in sequence space, enhancing the efficiency of laboratory evolution. In particular, molecular modeling methods provide a unique tool to grasp the sequence/structure/function relationship of the protein to evolve, with the only condition that a structural model is provided. With this book chapter, we tried to guide the reader through the state of the art of molecular modeling, discussing their strengths, limitations, and directions. In addition, we suggest a possible future template for in silico directed evolution where we underline two main points: a hierarchical computational protocol combining several different techniques and a synergic effort between simulations and experimental validation.

Peer Reviewed