基于遗传算法DYRK1A抑制剂的定量构效关系研究  

Quantitative Structure Activity Relationship of DYRK1A Inhibitor Based on Genetic Algorithm

作　　者：毕梦舟 张娜[1] BI Mengzhou;ZHANG Na(Faculty of Environment and Life,Beijing University of Technology,Beijing,100124 China)

机构地区：[1]北京工业大学环境与生命学部,北京100124

出　　处：《科技资讯》2022年第15期210-213,共4页Science & Technology Information

基　　金：2019年北京市自然科学基金面上资助《于新型直链骨架和药效碎片重组的靶向CK2抗癌抑制剂设计、优化及其活性评价》(项目编号:7192015)。

摘　　要：双特异性酪氨酸磷酸化调节激酶1A(Dual-Specificity Tyrosine Phosphorylation Regulated Kinase 1A,DYRK1A)被认定为重要的抗神经性退行性疾病的治疗靶点。该研究基于文献报道的117个DYRK1A杂环类抑制剂,采用遗传算法(Genetic Algorithm,GA)联合多元线性回归(Multiple Linear Regression,MLR)方法,建立了具有良好稳健性及预测能力的定量构效关系(Quantitative Structure-Activity Relationship,QSAR)模型,揭示了影响该类抑制剂活性的分子描述符包括SpAD_B(m),GATS5m,nCb-和B02[C-O],为新型DYRK1A抑制剂的开发提供理论基础和设计思路。Dual-Specificity Tyrosine Phosphorylation Regulated Kinase 1A(DYRK1A)has been considered as an important therapeutic target of neurodegenerative diseases.Based on 117 heterocyclic DYRK1A inhibitors reported in the literature,Genetic Algorithm(GA)combined with Multiple Linear Regression(MLR)method were used to develop a robust and predictive QSAR model.The optimal model shows that the molecular descriptors affecting DYRK1A inhibitory activity of include SpAD_B(m),GATS5m,nCb-and B02[C-O].All the results provide theoretical basis and optimization clues the development of novel DYRK1A inhibitors.

关 键 词：DYRK1A 杂环类抑制剂 定量结构-活性关系 遗传算法 多元线性回归 

分 类 号：O629.8[理学—有机化学]