EpiMix: an integrative tool for epigenomic subtyping using DNA methylation.

Area of Science:

• Epigenetics
• Genomics
• Computational Biology

Background:

• DNA methylation (DNAme) is a critical epigenetic regulator of gene expression, with aberrant patterns linked to complex diseases like cancer, immunological disorders, and cardiovascular conditions.
• Genome-wide DNAme quantification in large human cohorts is now feasible, yet existing analytical methods lack robust evaluation for detecting differential DNAme in heterogeneous patient populations.

Approach:

• Developed EpiMix, an end-to-end analytical framework for population-level DNA methylation and gene expression analysis.
• Enhanced EpiMix to analyze cis-regulatory elements in protein-coding genes, distal enhancers, and non-coding RNAs (microRNAs and lncRNAs).

Key Points:

• EpiMix demonstrates superior sensitivity in identifying differential DNAme, particularly in small patient subsets, compared to existing methods.
• Applied EpiMix to cell-type specific data, revealing novel epigenetic mechanisms in childhood food allergy.
• Discovered survival-associated, methylation-driven non-coding RNAs in non-small cell lung cancer.

Conclusions:

• EpiMix provides a powerful tool for population-level epigenetic analysis, improving the detection of disease-associated DNAme alterations.
• The framework facilitates the discovery of novel epigenetic regulators and biomarkers in diverse human diseases.