Epigenetic Clocks and DNA Methylation
1 Horvath 2013 — The Original Epigenetic Clock
Full Citation:: Horvath S. DNA methylation age of human tissues and cell types. Genome Biology. 2013;14(10):R115. (Horvath 2013) Key Contributions::
- Developed the first multi-tissue DNA methylation age predictor using 353 CpG sites
- Demonstrated that methylation age closely tracks chronological age across diverse tissue types
- Established the concept of "epigenetic age acceleration" as a biomarker of biological agingRelevance:: Foundation for the proposed ME/CFS-calibrated epigenetic clock (Speculation ME/CFS-Calibrated Epigenetic Clock). The Horvath methodology demonstrates that disease-state-specific methylation patterns can be quantified and tracked over time. Adapting this approach to ME/CFS-specific CpG sites would operationalize the consolidation variable \(\mathcal{M}\) from the formal causal hierarchy model. Certainty Assessment::
- *Quality:* Very high (foundational methodology paper; >14,000 citations)
- *Replication:* Extensively replicated across tissues, populations, and disease contexts
- *Certainty:* 0.95 (the epigenetic clock methodology is well-established; the application to ME/CFS is the novel, untested component)2 de Vega et al. 2014 — ME/CFS DNA Methylation Discovery
Full Citation:: de Vega WC, Vernon SD, McGowan PO. DNA methylation modifications associated with chronic fatigue syndrome. PLOS ONE. 2014;9(8):e104757. Vega, Vernon, and McGowan (2014) Key Contributions::
- First genome-wide methylation study in ME/CFS (Illumina 450K array)
- Identified 1,192 differentially methylated CpG sites in ME/CFS PBMCs
- Enrichment for immune function, cellular signaling, and metabolic regulation genesRelevance:: Provides the raw CpG site data needed to construct an ME/CFS-specific epigenetic clock. Used in Chapter Formal Causal Hierarchy Analysis to support the epigenetic clock hypothesis. Certainty Assessment::
- *Quality:* Moderate (small sample, $n = 49$; no independent replication cohort)
- *Replication:* Partially replicated by Trivedi 2018 and de Vega 2017/2021 with different cohorts
- *Certainty:* 0.50