生物催化进展:从计算到代谢工程（英文）

Advancements in biocatalysis:From computational to metabolic engineering

经过数次技术研究和超常创新战略的大发展,生物催化逐渐达到工业化水平,从而受到人们特别的关注.基于酶值,通过生物途径生产高附加值化合物和精细工业化学品成为人们最感兴趣的领域之一.更广泛的众多生物化学路线可由酶催化来实现,其中还有一些酶尚未被人们发现.另一方面,由于非同源底物和某些化学过程所必需的苛刻条件,导致酶催化过程的效率低、稳定性差,因而限制了生物催化的应用.因此,开发具有多催化特征、更高效率和稳定性的绿色催化剂,成为生物催化的重中之重.计算科学、代谢工程、合成生物,以及机器学习路线的运用为新催化剂的工程化提供了新方法.本文重点介绍了合成生物学和代谢工程在催化中的作用,讨论了用于催化的机器学习算法和如何选择一种预测蛋白质-配体相互作用的算法;为了预测键合和催化功能,综述了分子对接的重要性;最后给出了结束语、未来挑战和前景展望.

Through several waves of technological research and un‐matched innovation strategies,bio‐catalysis has been widely used at the industrial level.Because of the value of enzymes,methods for producing value‐added compounds and industrially‐relevant fine chemicals through biological methods have been developed.A broad spectrum of numerous biochemical pathways is catalyzed by enzymes,including enzymes that have not been identified.However,low catalytic efficacy,low stability,inhibition by non‐cognate substrates,and intolerance to the harsh reaction conditions required for some chemical processes are considered as major limitations in applied bio‐catalysis.Thus,the development of green catalysts with multi‐catalytic features along with higher efficacy and induced stability are important for bio‐catalysis.Implementation of computational science with metabolic engineering,synthetic biology,and machine learning routes offers novel alternatives for engineering novel catalysts.Here,we describe the role of synthetic biology and metabolic engineering in catalysis.Machine learning algorithms for catalysis and the choice of an algorithm for predicting protein‐ligand interactions are discussed.The importance of molecular docking in predicting binding and catalytic functions is reviewed.Finally,we describe future challenges and perspectives.