Identification of Multidimensional Regulatory Modules Through Multi-Graph Matching with Network Constraints.

Jiazhou Chen,Guoqiang Han,Aodan Xu,Hongmin Cai
DOI: https://doi.org/10.1109/tbme.2019.2927157
IF: 4.6
2019-01-01
IEEE Transactions on Biomedical Engineering
Abstract:OBJECTIVE:The accumulation of large amounts of multidimensional genomic data provides new opportunities to study multilevel biological regulatory associations. Identifying multidimensional regulatory modules (md-modules) from omics data is crucial to provide a comprehensive understanding of the regulatory mechanisms of biological systems.METHODS:We develop a multi-graph matching with multiple network constraints (MGMMNC) model to identify the md-modules. The MGMMNC model aims to accurately capture highly relevant md-modules by considering the relationships intra- and inter-multidimensional omics data, including interactions within a network and cycle consistency information. The proposed technique adopts a novel graph-smoothing similarity measurement for the highly contaminated genetic data.RESULTS:The superiority and effectiveness of MGMMNC have been demonstrated by comparative experiments with three state-of-the-art techniques using simulated and cervical cancer data.CONCLUSION:MGMMNC can accurately and efficiently identify the md-modules that are significantly enriched in gene ontology biological processes and in Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Many different level molecules in the same md-module collaboratively regulate the same pathway. Moreover, the md-modules are capable of stratifying patients into subtypes with significant survival differences.SIGNIFICANCE:The problem of identifying multidimensional regulatory modules from omics data is formulated as a multi-graph matching problem, and multiple network constraints and cycle consistency information are seamlessly integrated into the matching model.
What problem does this paper attempt to address?