Laboratory of Physical Chemistry and Chemical Thermodynamics, Faculty of Chemistry and Chemical Engineering, University of Maribor, Smetanova 17, SI-2000 Maribor, Slovenia.
Faculty of Mathematics, Natural Sciences and Information Technologies, University of Primorska, Glagoljaška 8, SI-6000 Koper, Slovenia.
Int J Mol Sci. 2021 Dec 30;23(1):393. doi: 10.3390/ijms23010393.
Since December 2019, the new SARS-CoV-2-related COVID-19 disease has caused a global pandemic and shut down the public life worldwide. Several proteins have emerged as potential therapeutic targets for drug development, and we sought out to review the commercially available and marketed SARS-CoV-2-targeted libraries ready for high-throughput virtual screening (HTVS). We evaluated the SARS-CoV-2-targeted, protease-inhibitor-focused and protein-protein-interaction-inhibitor-focused libraries to gain a better understanding of how these libraries were designed. The most common were ligand- and structure-based approaches, along with various filtering steps, using molecular descriptors. Often, these methods were combined to obtain the final library. We recognized the abundance of targeted libraries offered and complimented by the inclusion of analytical data; however, serious concerns had to be raised. Namely, vendors lack the information on the library design and the references to the primary literature. Few references to active compounds were also provided when using the ligand-based design and usually only protein classes or a general panel of targets were listed, along with a general reference to the methods, such as molecular docking for the structure-based design. No receptor data, docking protocols or even references to the applied molecular docking software (or other HTVS software), and no pharmacophore or filter design details were given. No detailed functional group or chemical space analyses were reported, and no specific orientation of the libraries toward the design of covalent or noncovalent inhibitors could be observed. All libraries contained pan-assay interference compounds (PAINS), rapid elimination of swill compounds (REOS) and aggregators, as well as focused on the drug-like model, with the majority of compounds possessing their molecular mass around 500 g/mol. These facts do not bode well for the use of the reviewed libraries in drug design and lend themselves to commercial drug companies to focus on and improve.
自 2019 年 12 月以来，新型 SARS-CoV-2 相关的 COVID-19 疾病已在全球范围内引发大流行并导致全球公共生活停摆。几种蛋白质已成为药物开发的潜在治疗靶标，我们寻求审查市售和已上市的针对 SARS-CoV-2 的文库，以进行高通量虚拟筛选 (HTVS)。我们评估了针对 SARS-CoV-2、以蛋白酶抑制剂为重点和以蛋白质-蛋白质相互作用抑制剂为重点的文库，以更好地了解这些文库的设计方式。最常见的方法是基于配体和结构的方法，以及使用分子描述符的各种过滤步骤。通常，这些方法结合使用以获得最终文库。我们认识到提供的靶向文库的丰富性，并赞赏包括分析数据；然而，必须提出严重的关切。即，供应商缺乏关于文库设计的信息，并且无法参考原始文献。在使用基于配体的设计时，也很少提供活性化合物的参考，通常仅列出蛋白质类或一般靶标面板，以及一般参考方法，例如基于结构的设计的分子对接。没有提供受体数据、对接协议，甚至没有参考应用的分子对接软件（或其他 HTVS 软件），也没有提供药效团或筛选设计细节。没有报告详细的官能团或化学空间分析，也没有观察到库针对共价或非共价抑制剂设计的特定方向。所有文库都包含 pan-assay interference compounds (PAINS)、快速消除无用化合物 (REOS) 和聚集剂，并且针对类药性模型，大多数化合物的分子量约为 500 g/mol。这些事实对于使用所审查的文库进行药物设计不利，并促使商业制药公司关注和改进。
Zotero 插件
