青岛大学学报（自然科学版）2025，Vol.38Issue(3)：57-61,117,6.DOI:10.3969/j.issn.1006-1037.2025.03.09
基于宿主—微生物网络的疾病预测算法
Disease Prediction Algorithm Based on Host-microbial Network
摘要
Abstract
An algorithm for disease prediction based on a host-microbe dual-layer network was proposed to address the issue of microbial interactions being overlooked in disease pre-diction methods.Host-microbe networks were constructed by microbial sequencing data and phylogenetic trees.The restart random walk algorithm was applied to propagate sig-nals across the network,and disease-associated microbial motifs were identified with prior microbial knowledge.These motifs were converted into predictive features and incorporat-ed into a model for disease prediction.The experimental results show that in the detection of colorectal cancer,the AUC value of logistic regression increases from 0.84 to 0.90,and the AUC value of random forest increases from 0.88 to 0.92.关键词
网络传播/生物标记物/疾病预测/宿主状态Key words
network propagation/microbial markers/disease prediction/host state分类
信息技术与安全科学引用本文复制引用
罗炜文,吴舜尧..基于宿主—微生物网络的疾病预测算法[J].青岛大学学报（自然科学版）,2025,38(3):57-61,117,6.基金项目
山东省自然科学基金(批准号:ZR2019PF012)资助 （批准号:ZR2019PF012）
山东省高等学校科技计划(批准号:J18KA356)资助. （批准号:J18KA356）
