Background: Traditional epigenetic sequencing approaches aggregate 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) into a single signal or misses 5hmC entirely, limiting discovery and mechanistic interpretation of regulatory biology. This limitation is especially consequential in genomic contexts where5hmC has known functional roles, including at enhancers, where it marks activation, gene bodies, and is a stable mark of tissue-specific actively transcribed genes.
Methods: We used 6-base sequencing data to independently quantify 5mC and 5hmC across cfDNA and tissue datasets. We assessed biological signal using correlation and principal component analysis, identified 5mC and 5hmC-specific differentially methylated regions (DMRs), examined modification trajectories across disease stages, and evaluated functional interpretation through genomic feature mapping, sample-level feature extraction, and downstream gene ontology and gene set enrichment analyses.
Results: Independent quantification of 5mC and 5hmC revealed biologically meaningful differences that were obscured in conventional methylation analyses. We identified hidden DMRs in which opposing 5mC and 5hmC changes neutralized the aggregate signal, as well as regions that appeared hypermethylated by conventional analysis but were instead driven by 5hmC gains consistent with activation rather than repression. In samples across disease stages, early 5hmC signal preceded later 5mC loss, supporting a trajectory-based model of epigenetic change and enabling earlier detection of gene activation. Functional interpretation of these DMRs highlighted activation-linked states, including promoter-associated 5mC loss and first-exon 5hmC gain in relevant genes, while feature extraction and heatmap-based analyses clearly separated samples using 6-base features. Gene ontology and gene set enrichment analyses further linked these patterns to interpretable biological pathways.
Conclusion: These analyses show that 6-base DMR analysis recovers functional genomic insight and reveals regulatory biology that remains invisible to conventional approaches. Resolving 5mC and 5hmC improves discovery of novel biomarkers and enables mechanistic interpretation of differential methylation, with clear value for downstream classification and biological discovery.
