A generalized higher-order correlation analysis framework for multi-omics network inference

• DOI: 10.1371/journal.pcbi.1011842
• Publication Date: 2025-04-14T20:05:09Z
• Authors: Weixuan Liu; Katherine A. Pratte; Peter J. Castaldi; Craig Hersh; Russell P. Bowler; Farnoush Banaei-K...; Katerina J. Kechris
• Keywords: Omics

Multiple -omics (genomics, proteomics, etc.) profiles are commonly generated to gain insight into a disease or physiological system. Constructing multi-omics networks with respect the trait(s) of interest provides an opportunity understand relationships between molecular features but integration is challenging due multiple data sets high dimensionality. One approach use canonical correlation integrate one two omics types and single trait interest. However, these methods may be limited (1) not accounting for higher-order correlations existing among features, (2) computational inefficiency when extending more than using penalty term-based sparsity method, (3) lack flexibility focusing on specific (e.g., omics-to-phenotype versus omics-to-omics correlations). In this work, we have developed novel network analysis pipeline called Sparse Generalized Tensor Canonical Correlation Analysis Network Inference (SGTCCA-Net) that can effectively overcome limitations. We also introduce implementation improve summarization downstream analyses. Simulation real-data experiments demonstrate effectiveness our method inferring