Suppr超能文献

利用主成分分析和人工神经网络相结合的策略探究植物酚类化合物抗氧化活性的分子要求。

Probing into the Molecular Requirements for Antioxidant Activity in Plant Phenolic Compounds Utilizing a Combined Strategy of PCA and ANN.

作者信息

Agatonovic-Kustrin Snezana, Morton David W, Ristivojevic Petar

机构信息

School of Pharmacy and Applied Science, La Trobe Institute for Molecular Sciences, La Trobe University, Edwards Rd, Bendigo 3550. Australia.

Innovation Centre of the Faculty of Chemistry Ltd., Studentski trg 12-16, 11000 Belgrade. Serbia.

出版信息

Comb Chem High Throughput Screen. 2017;20(1):25-34. doi: 10.2174/1386207320666170102123146.

Abstract

AIM AND OBJECTIVE

This study investigates molecular structural requirements that are responsible for the antioxidant activity in phenolic compounds.

METHOD

Antioxidant activity of compounds was determined with a 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical assay. Principal component analysis (PCA) was used to classify phenolic antioxidants according to the key molecular features that contribute to their antioxidant activity. Artificial neutral networks (ANNs) was used to develop a predictive QSAR model.

RESULTS

Both models agreed that structural characteristics of phenolic compounds responsible for the antioxidant activity include: (1) number and position of alcohol groups on the aromatic ring; (2) molecular size; (3) flexibility/bulkiness; and (4) water solubility. PCA has classified data into phenolic acids and flavonoids, suggesting two distinct mechanisms of action. ANN has confirmed different mechanisms of action for flavonoids and polyphenolic acids, i.e. breaking of free radical chain reactions by donation of a hydrogen atom to neutralise a free radical and the chelating ability of polyphenolic acids.

CONCLUSION

Although two phenolic acids may have the same relative polarity, their different functional groups may drastically change the nature of their interactions with free radicals, and their antioxidant activity.

摘要

目的

本研究调查了酚类化合物中负责抗氧化活性的分子结构要求。

方法

用2,2-二苯基-1-苦基肼(DPPH)自由基测定法测定化合物的抗氧化活性。主成分分析(PCA)用于根据有助于酚类抗氧化剂抗氧化活性的关键分子特征对其进行分类。人工神经网络(ANNs)用于建立预测性定量构效关系(QSAR)模型。

结果

两种模型均认为,负责抗氧化活性的酚类化合物的结构特征包括:(1)芳环上醇基的数量和位置;(2)分子大小;(3)柔韧性/体积;以及(4)水溶性。PCA已将数据分为酚酸和黄酮类,表明有两种不同的作用机制。ANN已证实黄酮类和多酚酸有不同的作用机制,即通过提供氢原子中和自由基来打破自由基链反应以及多酚酸的螯合能力。

结论

尽管两种酚酸可能具有相同的相对极性,但它们不同的官能团可能会极大地改变它们与自由基相互作用的性质及其抗氧化活性。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验