Disease prediction by network information gain on a single sample basis  

作　　者：Jinling Yan Peiluan Li Ying Li Rong Gao Cheng Bi Luonan Chen 

机构地区：[1]School of Mathematics and Statistics,Henan University of Science and Technology,Luoyang 471023,China [2]Key Laboratory of Information Fusion Technology of Ministry of Education,School of Automation,Northwestern Polytechnical University,Xi'an 710072,China [3]Longmen Laboratory,Luoyang 471003,China [4]Key Laboratory of Systems Biology,Center for Excellence in Molecular Cell Science,Shanghai Institute of Biochemistry and Cell Biology,Chinese Academy of Sciences,Shanghai 200031,China [5]Key Laboratory of Systems Health Science of Zhejiang Province,Hangzhou Institute for Advanced Study,University of Chinese Academy of Sciences,Chinese Academy of Sciences,Hangzhou 310024,China [6]Guangdong Institute of Intelligence Science and Technology,Zhuhai 519031,China [7]School of Life Science and Technology,ShanghaiTech University,Shanghai 201210,China

出　　处：《Fundamental Research》2025年第1期215-227,共13页自然科学基础研究(英文版)

基　　金：supported by the National Natural Science Foundation of China(31930022,12131020,T2341007,T2350003);National Key R&D Program of China(2022YFA1004800);Strategic Priority Research Program of the Chinese Academy of Sciences(XDB38040400);Special Fund for Science and Technology Innovation Strategy of Guangdong Province(2021B0909050004,2021B0909060002);Key-Area Research and Development Program of Guangdong Province(2021B0909060002);MajorKey Project of PCL(PCL2021A12),and JST Moonshot R&D(JPMJMS2021).

摘　　要：There are critical transition phenomena during the progression of many diseases.Such critical transitions are usually accompanied by catastrophic disease deterioration,and their prediction is of significant importance for disease prevention and treatment.However,predicting disease deterioration solely based on a single sample is a difficult problem.In this study,we presented the network information gain(NIG)method,for predicting the critical transitions or disease state based on network flow entropy from omics data of each individual.NIG can not only efficiently predict disease deteriorations but also detect their dynamic network biomarkers on an individual basis and further identify potential therapeutic targets.The numerical simulation demonstrates the effectiveness of NIG.Moreover,our method was validated by successfully predicting disease deteriorations and identifying their potential therapeutic targets from four real omics datasets,i.e.,an influenza dataset and three cancer datasets.

关 键 词：Tipping point Network information gain Dynamic network biomarker Disease prediction Network flow entropy Drug target 

分 类 号：R73[医药卫生—肿瘤]