Computational Chemistry Department, Institute of Chemistry Timisoara of the Romanian Academy, B-dul Mihai Viteazu 24, RO-300223, Timisoara, Romania.
Natural Science Laboratory, Toyo University, 5-28-20 Hakusan, Bunkyo-ku, Tokyo, 112-8606, Japan.
Mol Inform. 2019 Aug;38(8-9):e1800119. doi: 10.1002/minf.201800119. Epub 2019 Jan 11.
Neonicotinoids are known to have high insecticidal potency, low mammalian toxicity and relatively tough activity for the development of resistance against aphids. A series of guadipyr insecticides, active against Myzus persicae was engaged in silico studies, based on Multiple Linear Regression (MLR), Partial Least Squares regression (PLS), Artificial Neural Networks (ANN), Support Vector Machine (SVM) and Pharmacophore modeling. Robust and predictive models were built using correlations between the insecticidal profile, expressed by experimental pLC values, and molecular descriptors, calculated from the energy optimized structures. Four new potential insecticides active against Myzus persicae and their predicted pLC toxicity values were reported for the first time. The models presented here can be used as an approach in the screening and prioritization of chemicals in a scientific and regulatory frame and for toxicity prediction.
新烟碱类杀虫剂具有杀虫活性高、哺乳动物毒性低、对蚜虫抗性发展相对较强的特点。本研究基于多元线性回归（MLR）、偏最小二乘回归（PLS）、人工神经网络（ANN）、支持向量机（SVM）和药效团模型，对一系列具有杀蚜活性的呱啶类杀虫剂进行了计算机模拟研究。利用由能量优化结构计算得到的分子描述符与实验测定的 pLC 值之间的相关性，建立了稳健且具有预测能力的模型。首次报道了 4 种新型具有杀蚜活性的潜在杀虫剂及其预测的 pLC 毒性值。本文提出的模型可用于在科学和监管框架内筛选和优先考虑化学品，并进行毒性预测。
