Abstract：
Many viruses, including SARS-CoV-2, the coronavirus responsible for the COVID-19 pandemic, enter host cells through a process of cell-viral membrane fusion that is activated by proteolytic enzymes. Typically, these enzymes are host cell proteases. Identifying the proteases that activate the virus is not a simple task but is important for the development of new antiviral drugs. In this study, we developed a bioinformatics method for identifying proteases that can cleave viral envelope glycoproteins. The proposed approach involves the use of predictive models for the substrate specificity of human proteases and the application of a structural analysis method for predicting the vulnerability of protein regions to proteolysis based on their 3D structures. Specificity models were constructed for 169 human proteases using information on their known substrates. A previously developed method for structural analysis of potential proteolysis sites was applied in parallel with specificity models. Validation of the proposed approach was performed on the SARS-CoV-2 spike protein, whose proteolysis sites have been well studied.
摘要：
许多病毒,包括SARS-CoV-2，这是导致COVID-19大流行的冠状病毒，通过蛋白水解酶激活的细胞-病毒膜融合过程进入宿主细胞。通常,这些酶是宿主细胞蛋白酶。鉴定激活病毒的蛋白酶不是一项简单的任务，但对于开发新的抗病毒药物很重要。在这项研究中,我们开发了一种生物信息学方法来鉴定可以切割病毒包膜糖蛋白的蛋白酶。所提出的方法涉及使用用于人类蛋白酶的底物特异性的预测模型，以及基于其3D结构预测蛋白质区域对蛋白质水解的脆弱性的结构分析方法的应用。使用有关其已知底物的信息，为169个人蛋白酶构建了特异性模型。先前开发的用于潜在蛋白水解位点的结构分析的方法与特异性模型平行应用。对SARS-CoV-2刺突蛋白进行了拟议方法的验证，其蛋白水解位点已被充分研究。
