Multi-View Clustering for Integration of Gene Expression and Methylation Data With Tensor Decomposition and...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• DNA methylation and gene expression are key epigenetic patterns for understanding biological mechanisms.
• Existing integrative algorithms struggle with data heterogeneity and intrinsic relationships between omics data.
• This heterogeneity limits the accurate exploitation of epigenetic patterns.

Purpose of the Study:

• To propose a novel multi-view clustering algorithm (MCSL-LTC) for integrating gene expression and DNA methylation data.
• To address the limitations of current methods in handling data heterogeneity.
• To enhance the accuracy and consistency of integrated genomic data analysis.

Main Methods:

• Developed a multi-view clustering approach with self-representation learning and low-rank tensor constraints (MCSL-LTC).
• Treated gene expression and DNA methylation as complementary views.
• Learned low-dimensional features and fused them in a unified tensor space with low-rank constraints to capture complementary information and avoid heterogeneity.

Main Results:

• MCSL-LTC effectively captures complementary information between different omics data views.
• The proposed method avoids the heterogeneity issue inherent in omics data.
• Experimental results show MCSL-LTC outperforms state-of-the-art methods in accuracy on both social and cancer datasets.

Conclusions:

• MCSL-LTC provides an effective and efficient method for integrating heterogeneous genomic data.
• The approach enhances the consistency and accuracy of epigenetic pattern analysis.
• This facilitates a deeper understanding of underlying biological system mechanisms.