Scholarlink

The integration of machine learning (ML) and prediction of metabolic pathways has enhanced the identification of drug-target interactions (DTIs) and hence effective drug discovery. Traditional drug discovery methods, which are often time-consuming and costly, are being supplemented by ML algorithms capable of analyzing large biological datasets and revealing complex patterns. Deep learning (DL), a subset of ML, has emerged as a powerful tool for modeling DTIs, leveraging neural networks to predict potential drug candidates based on structural and chemical properties (such as DeepDTIs, DeepACTION, DeepPurpose, and NeoDTI). We emphasize the integration of diverse data sources, including genomic, proteomic, and chemical information, utilizing platforms like DeepChem and AutoML to train robust predictive models. Furthermore, we discuss the implications of accurate DTI and metabolic pathways predictions on the efficiency of drug design and development, significantly reducing the time and costs associated with traditional methods. Our findings underscore the potential of machine learning to accelerate the discovery of novel drugs and optimize existing therapeutic strategies.

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

Machine Learning Approaches in Metabolic Pathway Predictions and Drug-Target Interactions: Advancing Drug Discovery

  • 章节

  • 发表日期：

    2025年

查看全文
  • Mohamed E. Hasan,
  • Rawhia R. Allam,
  • Alaa A. Hemeida,
  • Mahmoud ElHefnawi

摘要

The integration of machine learning (ML) and prediction of metabolic pathways has enhanced the identification of drug-target interactions (DTIs) and hence effective drug discovery. Traditional drug discovery methods, which are often time-consuming and costly, are being supplemented by ML algorithms capable of analyzing large biological datasets and revealing complex patterns. Deep learning (DL), a subset of ML, has emerged as a powerful tool for modeling DTIs, leveraging neural networks to predict potential drug candidates based on structural and chemical properties (such as DeepDTIs, DeepACTION, DeepPurpose, and NeoDTI). We emphasize the integration of diverse data sources, including genomic, proteomic, and chemical information, utilizing platforms like DeepChem and AutoML to train robust predictive models. Furthermore, we discuss the implications of accurate DTI and metabolic pathways predictions on the efficiency of drug design and development, significantly reducing the time and costs associated with traditional methods. Our findings underscore the potential of machine learning to accelerate the discovery of novel drugs and optimize existing therapeutic strategies.