Microbiome Data Analysis by Symmetric Non-negative Matrix Factorization With Local and Global Regularization

摘要

Network is an efficient tool to organize complicated data. Laplacian graph has attracted more and more attentions for its good property and been applied many tasks including clustering, feature selection and so on. Recently, some studies indicate that though Laplacian can captures the global information of data, it lacks the power to capture fine-grained structure inherent in network. In contrast, Vicus matrix can make full use of local topologies information of the data. Based on this consideration, in this paper we simultaneously introduce Laplacian and Vicus graphs into symmetric nonnegative matrix factorization framework (LVSNMF) to seek and exploit the global and local structure patterns existed in the original data. Extensive experiments are conducted on three real datasets (cancer, cell populations and microbiome data). The experimental results show the proposed LVSNMF algorithm significantly outperforms other competing algorithms, suggesting its potential in biological data analysis.