NMR metabolomics-guided DNA methylation mortality predictors.

Area of Science:

• Biomarkers
• Genomics
• Metabolomics

Background:

• ¹H-NMR metabolomics and DNA methylation are known age-related mortality biomarkers.
• These biomarkers provide independent signals for mortality and frailty.

Purpose of the Study:

• To investigate if metabolomic mortality signals can guide the development of DNA methylation-based mortality predictors.
• To leverage multi-omics data from population studies to build novel mortality predictors.

Main Methods:

• Trained DNA methylation-based surrogates for 64 metabolomic analytes using multi-omics data from 5238 individuals.
• Utilized DNA methylation assays to reconstruct specific metabolic analytes and a known mortality risk score (MetaboHealth).

Main Results:

• Accurate reconstruction of analytes related to inflammation, fluid balance, and lipid metabolism.
• Sixteen derived surrogates, including MetaboHealth, showed significant independent associations with mortality.
• MetaboHealth surrogate was accurately reconstructed from DNA methylation data.

Conclusions:

• Metabolic analyte-derived surrogates derived from DNA methylation show potential as valuable mortality signals.
• These surrogates, when added to epigenetic clocks like GrimAge, enhance mortality prediction.