Scholarlink

<p>ALOX15 is a key regulatory enzyme in multiple pathological processes including inflammation, cancer, and cardiovascular disease, rendering the development of potent inhibitors of this enzyme of significant clinical importance. This study aims to screen and optimise novel ALOX15 inhibitors by integrating multiple computational chemistry and rational drug design approaches. First, we constructed a 3D-QSAR pharmacophore model based on 28 known inhibitors to screen the Comprehensive Marine Natural Products Database (CMNPD), which contained 25,224 compounds. Combining machine learning and molecular docking methods, we preliminarily identified three molecules. Through ADMET analysis and scaffold hopping optimisation, we obtained three candidate compounds exhibiting both high binding affinity and favourable pharmacokinetic properties. Toxicity predictions indicated all compounds fell within the confidence interval for predicted non-toxicity or low toxicity. Molecular dynamics simulations further confirmed strong binding affinity between the candidates and ALOX15. Finally, off-target analysis identified two novel ALOX15 inhibitors, providing potential candidate molecules for subsequent development of ALOX15-targeted therapeutics.</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

Harnessing machine learning and multi-scale modeling to discover novel ALOX15 inhibitors from marine natural products

  • 期刊论文

  • 发表日期：

    2026年4月21日

查看全文
  • Xiyi Zheng,
  • Xiaomin Lin,
  • Jiahua Tao,
  • Lianxiang Luo

摘要

ALOX15 is a key regulatory enzyme in multiple pathological processes including inflammation, cancer, and cardiovascular disease, rendering the development of potent inhibitors of this enzyme of significant clinical importance. This study aims to screen and optimise novel ALOX15 inhibitors by integrating multiple computational chemistry and rational drug design approaches. First, we constructed a 3D-QSAR pharmacophore model based on 28 known inhibitors to screen the Comprehensive Marine Natural Products Database (CMNPD), which contained 25,224 compounds. Combining machine learning and molecular docking methods, we preliminarily identified three molecules. Through ADMET analysis and scaffold hopping optimisation, we obtained three candidate compounds exhibiting both high binding affinity and favourable pharmacokinetic properties. Toxicity predictions indicated all compounds fell within the confidence interval for predicted non-toxicity or low toxicity. Molecular dynamics simulations further confirmed strong binding affinity between the candidates and ALOX15. Finally, off-target analysis identified two novel ALOX15 inhibitors, providing potential candidate molecules for subsequent development of ALOX15-targeted therapeutics.