Scholarlink

<p>Experiments were conducted to evaluate the antifungal activity of chain aldehydes with potent antifungal properties against <i>Penicillium italicum</i> (<i>P. italicum</i>) resistant to prochloraz. The results showed that <i>trans,trans</i>-2,4-nonadienal and <i>trans,trans</i>-2,4-octadienal possessed the strong antifungal activity with MIC values of 0.10 μL/mL for both compounds. Density functional theory (DFT) revealed that compounds with highly delocalized electron systems exhibited significantly stronger antifungal potency. Machine learning (ML) techniques were used to build regression models for identifying the influential structure features of aldehydes, and the random forest (RF) model had lower error levels. We can conclude that the DFT coupled with ML is a reliable, fast and inexpensive tool for solving problems related to the structure properties and antifungal activity. The strategy can be extended to find safe and effective novel inhibitors of <i>P. italicum</i> resistant to prochloraz for the development of green inhibitors of postharvest diseases in fruits and vegetables.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Density functional theory and machine learning approaches for deciphering the antifungal properties of aldehydes against Penicillium italicum resistant to prochloraz

  • 期刊论文

  • 发表日期：

    2026年6月29日

查看全文
  • Chunjing Wang,
  • Xingyu Liu,
  • Xiumei Chen,
  • Jinxin Che,
  • Nengguo Tao,
  • Fei Pan

摘要

Experiments were conducted to evaluate the antifungal activity of chain aldehydes with potent antifungal properties against Penicillium italicum (P. italicum) resistant to prochloraz. The results showed that trans,trans-2,4-nonadienal and trans,trans-2,4-octadienal possessed the strong antifungal activity with MIC values of 0.10 μL/mL for both compounds. Density functional theory (DFT) revealed that compounds with highly delocalized electron systems exhibited significantly stronger antifungal potency. Machine learning (ML) techniques were used to build regression models for identifying the influential structure features of aldehydes, and the random forest (RF) model had lower error levels. We can conclude that the DFT coupled with ML is a reliable, fast and inexpensive tool for solving problems related to the structure properties and antifungal activity. The strategy can be extended to find safe and effective novel inhibitors of P. italicum resistant to prochloraz for the development of green inhibitors of postharvest diseases in fruits and vegetables.