摘要
针对现有代谢路径预测算法仅以单一代谢反应为特征进行路径排序,难以得到高质量路径预测结果的问题,本文引入代谢路径关键反应段的概念,用以捕捉和度量代谢路径的结构功能特征,为充分利用路径的结构功能特征信息,提出一种融合代谢路径关键反应段的信息熵与互信息对代谢路径进行评分和排序的模型,在此基础上设计一种融合关键反应段特征信息的代谢路径排序方法KPRank.在KEGG代谢数据库上的实验结果表明,本文方法KPRank有效提高了代谢路径预测结果的质量.
Aiming at the problem that the existing metabolic pathway prediction algorithms only sort the predicted pathways based on single reaction,which makes it difficult to obtain high quality resulting pathways.This paper introduces the concept of key reaction set to capture and measure the structural function feature of metabolic pathways.In order to make use of the structural function feature of metabolic pathways,this paper proposes a model that incorporates the information entropy and mutual information of key reaction set of metabolic pathways into scoring and sorting metabolic pathways and designs a method for sorting metabolic pathway called KPRank.The experiments on the KEGG metabolic database show that,our method KPRank can effectively improve the quality of the pathways produced by the existing metabolic pathway prediction algorithms.
作者
谢雨丝
卢吉晓
黄毅然
钟诚
XIE Yu-si;LU Ji-xiao;HUANG Yi-ran;ZHONG Cheng(School of Computer and Electronics and Information,Guangxi University,Nanning 530004,China;Radiotherapy Technology Center,Guangxi Medical University Cancer Hospital,Nanning 530021,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2023年第8期1672-1679,共8页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61862006,61861004)资助
广西自然科学基金项目(2020GXNSFAA159074)资助。
关键词
代谢路径排序
关键反应段
信息熵
互信息
metabolic pathway sorting
key reaction set
information entropy
mutual information