Faculty of Natural Sciences, Department of Chemistry, University of Oriente, 90500 Santiago de Cuba, Cuba.
Mol Divers. 2011 Nov;15(4):901-9. doi: 10.1007/s11030-011-9320-7. Epub 2011 Jun 2.
The increasing resistance of several phytopathogenic fungal species to the existing agrochemical fungicides has alarmed to the worldwide scientific community. There is no available methodology to predict in an efficient way if a new fungicide will have resistance risk due to fungal species which cause considerable crop losses. In an attempt to overcome this problem, a multi-resistance risk QSAR model, based on substructural descriptors was developed from a heterogeneous database of compounds. The purpose of this model is the classification, design, and prediction of agrochemical fungicides according to resistance risk categories. The QSAR model classified correctly 85.11% of the fungicides and the 85.07% of the inactive compounds in the training series, for an accuracy of 85.08%. In the prediction series, the percentages of correct classification were 85.71 and 86.55% for fungicides and inactive compounds, respectively, with an accuracy of 86.39%. Some fragments were extracted and their quantitative contributions to the fungicidal activity were calculated taking into consideration the different resistance risk categories for agrochemical fungicides. In the same way, some fragments present in molecules with fungicidal activity and with negative contributions were analyzed like structural alerts responsible of resistance risk.
几种植物病原真菌对现有农用杀菌剂的抗药性不断增强，这一现象引起了全球科学界的警觉。目前还没有一种有效的方法可以预测一种新的杀菌剂是否会因引起严重作物损失的真菌物种而具有抗药性风险。为了克服这一问题，我们从一个异构化合物数据库中开发了一个基于亚结构描述符的多抗性风险 QSAR 模型。该模型的目的是根据抗性风险类别对农用杀菌剂进行分类、设计和预测。QSAR 模型正确地对训练系列中的 85.11%的杀菌剂和 85.07%的非活性化合物进行了分类，准确率为 85.08%。在预测系列中，杀菌剂和非活性化合物的正确分类百分比分别为 85.71%和 86.55%，准确率为 86.39%。提取了一些片段，并考虑了农用杀菌剂的不同抗性风险类别，计算了它们对杀菌活性的定量贡献。同样，分析了具有杀菌活性和负贡献的分子中存在的一些片段，这些片段可能是导致抗药性风险的结构警报。
