LMSM

LMSM identifies lncRNA-related miRNA sponge modules by integrating heterogeneous molecular data to characterize sponge-mediated regulatory modules in breast cancer (BRCA) using datasets such as The Cancer Genome Atlas (TCGA).

Key Features:

• Integration of heterogeneous data: Utilizes gene expression data alongside other relevant datasets to evaluate the influence of shared miRNAs on clustered sponge lncRNAs and mRNAs.
• Application to BRCA (TCGA): Applied to human breast cancer (BRCA) datasets from The Cancer Genome Atlas (TCGA) to identify disease-related modules and subtype-specific modules.
• Identification of mediating miRNAs: Highlights mediating miRNAs that act as crosslinks across multiple identified sponge modules.
• Statistical significance assessment: Reports that identified LMSM modules are statistically significant, supporting robustness of the findings.
• miRNA target prediction: Predicts miRNA targets as part of the module analysis to expand regulatory network mapping.
• Improved subtype classification: Multi-label classification analysis indicates LMSM modules outperform baseline methods in classifying BRCA subtypes.
• Comparison with graph clustering: Outperforms traditional graph clustering-based strategies in identifying breast cancer-related sponge lncRNA modules from heterogeneous data.

Scientific Applications:

• BRCA module discovery: Identification of lncRNA-related miRNA sponge modules implicated in breast cancer and its subtypes.
• Modular regulatory mechanism analysis: Investigation of modular regulatory mechanisms of sponge lncRNAs and their mediated interactions with mRNAs via shared miRNAs.
• Regulatory network mapping: Expansion of miRNA target prediction and mapping of sponge-mediated regulatory networks.
• Biomarker and therapeutic target identification: Support for identifying candidate biomarkers and therapeutic targets related to sponge lncRNA modules.
• Extension to other diseases: Applicable to studying diseases where miRNA sponging contributes to regulatory mechanisms beyond breast cancer.

Methodology:

Integrates gene expression and other heterogeneous datasets, applies the framework to TCGA BRCA data, identifies mediating miRNAs and statistically significant modules, performs miRNA target prediction, and evaluates module performance using multi-label classification and comparisons with graph clustering-based strategies.