基于分子指纹对黄酮类化合物及其抗DPPH自由基活性的QSAR分析

应用定量构效关系（Quantitative structure activity relationship, QSAR）研究阐明黄酮类化合物（Flavonoid compounds, FCs）的子结构指纹（Substructure fingerprint）与1,1-二苯基-2-三硝基苯肼（1,1-Diphenyl-2-picrylhydrazyl, DPPH）自由基清除能力之间的关系，从而指导高效抗氧化物质的设计和发现。在PubMed数据库中收集77个具有明确抗氧化活性的黄酮类化合物，而在ChEMBL数据库中收集86个无抗DPPH活性的黄酮类化合物。这163个黄酮类化合物的子结构指纹由PubChem系统生成，然后通过卡方检验筛选出与黄酮类化合物的抗氧化活性显著相关的分子指纹，最后通过判别分析建立预测QSAR模型，并采用回代法和交叉验证法对已建立的模型进行准确性和稳健性的验证。结果表明，黄酮类化合物抗DPPH自由基活性与ESSSR环的计数、简单相邻原子的类型和简单的SMARTS模式等因素有关。此外，所建立的QSAR模型能较好地预测黄酮类化合物的DPPH自由基清除活性，可用于评价候选抗氧化剂的潜力。

QSAR analysis of flavonoid compounds and their anti-DPPH radical activity based on substructure fingerprints

QSAR study was applied to elucidate the relationship between the substructure fingerprints of flavonoid compounds (FCs) and their DPPH radical scavenging capability, to guide the design and discovery of highly-potent antioxidant substances more efficiently. PubMed database was used to collect 77 FCs with antioxidant activity, whereas, 86 negative FCs were found in ChEMBL database. The substructure fingerprints of these 163 FCs were generated with PubChem System, and test was employed to filter the substructure fingerprints significantly related to the antioxidant activity of FCs. The prediction QSAR model was then established by discriminant analysis, and the obtained model was finally verified by the back-substitution and Leave-One-Out cross-validation methods. It was revealed that the anti-DPPH radical activity of FCs was correlated with Rings in a canonic Extended Smallest Set of Smallest Rings (ESSSR) ring set, Simple atom nearest neighbors and Simple SMARTS patterns. The established QSAR model could explicitly predict the antioxidant activity of FCs, thus were applicable to evaluate the potential of candidates as antioxidants.