AUTHOR=Zhao Yongbiao , Ma Yuanyuan , Zhang Qilin 

TITLE=Metabolite-disease interaction prediction based on logistic matrix factorization and local neighborhood constraints

JOURNAL=Frontiers in Psychiatry

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2023.1149947

DOI=10.3389/fpsyt.2023.1149947

ISSN=1664-0640

ABSTRACT=More and more evidences show that metabolites are closely related to human diseases. Identifying disease-related metabolites is especially important for the diagnosis and treatment of disease. The previous works mainly focus on the global topological information of metabolite and disease similarity networks, may ignore the local tiny structural of metabolites and diseases, which may lead to the insufficiency and inaccuracy on the task of metabolite-disease potential interactions mining. To solve this problem, in this manuscript we propose a novel metabolite-disease interaction prediction method with logical matrix factorization and local nearest neighborhood constraints (LMFLNC). The proposed LMFLNC method can well preserve the geometrical structure of original data and can effectively predict the underlying associations between metabolites and diseases. Firstly, the algorithm constructed metabolite-metabolite and disease-disease similarity networks by integrating multi-source heterogeneous omics data of microbiome. Then, the local spectral matrices based on these two networks were established and used as the input of the model together with the known metabolite-disease interaction matrix. Finally, the probability of metabolite-disease interaction was calculated according to the latent representation matrices of metabolites and diseases. Extensive experiments on the metabolite-disease interaction data were conducted, the experimental results show that the proposed method outperforms the second-best algorithm than 5.28%/5.61% in terms of AUPR and F1, respectively, which demonstrates the effectiveness of the proposed LMFLNC on the task of metabolite-disease interaction predictions.