一个基于联体的新型卷积神经网络,用于在药物发现中识别P-糖蛋白联体
Mary Margarat Valentine A Neela1,2, Subbarao Peram3
1Computer Science & Engineering, Vignan's foundation for Science, Technology and Research (Deemed to be University), Guntur, India. marym.neela@gmail.com.
一种新的深度学习模型准确地预测P-glycoprotein (P-gp) 基质,克服了以前方法的局限性. 这种计算工具通过改善P-gp相互作用预测来增强药物发现和个性化医疗.
科学领域:
- 计算药理学计算药理学
- 药物发现 药物发现 药物发现
- 生物信息学是一种生物信息学.
背景情况:
- P-glycoprotein (P-gp) 是一个关键的载体,涉及抗药性和药物输送和癌症治疗的挑战.
- 预测P-gp相互作用的现有计算方法受到小数据集的限制.
研究的目的:
- 开发一种新的,高度准确的计算框架,用于分类和预测P-gp基质.
- 改进现有的P-gp相互作用预测模型.
主要方法:
- 一个基于新基的卷积神经网络 (NLCNN) 框架的开发.
- 在197个P-gp基质的精选数据集上训练NLCNN.
- 分子对接和基于联体的深度学习技术的整合.
- 使用人类P-gp的同质模型进行对接分析.
主要成果:
- 该NLCNN框架实现了80%的平均预测准确度.
- 与传统的卷积神经网络 (CNN) 和支持矢量机器 (SVM) 相比,该模型的精度和回忆率提高了19-24%.
- 通过高斯的RBF内核,NLCNN的性能优于SVM.
结论:
- 拟议的NLCNN框架为预测P-gp抑制剂和基质提供了一个强大,准确和简单的工具.
- 这一进步对药物发现和个性化医学有重大影响,因为它能够精确地预测P-gp相互作用.
更多相关视频
09:03Parallel Interrogation of β-Arrestin2 Recruitment for Ligand Screening on a GPCR-Wide Scale using PRESTO-Tango Assay
Published on: March 10, 2020
08:31Biosensor-based High Throughput Biopanning and Bioinformatics Analysis Strategy for the Global Validation of Drug-protein Interactions
Published on: December 1, 2020
相关概念视频
G Protein-coupled Receptors
GPCRs are also called heptahelical, 7TM, or serpentine receptors, and consist of seven (H1-H7) transmembrane alpha-helices that span the bilayer to form a cylindrical core. The transmembrane helices are connected by three extracellular loops and three...
Ligand Binding Sites
Protein-ligand interactions are quite specific; even though numerous potential ligands surround a cellular protein at any given time, only a particular ligand can bind to that protein. Moreover, a ligand binds only to a dedicated area on the surface of the protein, known as the...
Transducer Mechanism: G Protein–Coupled Receptors
GPCRs are also called heptahelical,...
Protein-protein Interfaces
G-protein Coupled Receptors
