Building Minimized Epigenetic Clock by iPlex MassARRAY Platform.

Area of Science:

• Epigenetics
• Computational Biology
• Gerontology

Background:

• Epigenetic clocks estimate biological and chronological age using DNA methylation.
• Genome-wide approaches are comprehensive but costly.
• Targeted methods offer efficiency for epigenetic clock construction.

Purpose of the Study:

• To assess the feasibility of a minimized epigenetic clock using iPlex MassARRAY technology.
• To evaluate age prediction accuracy compared to established methods.
• To explore machine learning models for clock development.

Main Methods:

• Collected DNA methylation data from eight CpG sites in specific genes using iPlex MassARRAY and Illumina EPIC array.
• Applied machine learning models (linear, DNN, GBDT) including TabNet.
• Compared clock performance against established epigenetic clocks.

Main Results:

• Five CpG sites showed significant age correlation and good data concordance between technologies.
• Three CpG sites had weak age correlation and low concordance.
• The TabNet model achieved the best performance with a Mean Absolute Error (MAE) of 5.99.
• The minimized clock showed slightly higher age prediction errors than existing clocks.

Conclusions:

• A minimized epigenetic clock using targeted methylation data is feasible.
• iPlex MassARRAY technology can be utilized for targeted epigenetic clock construction.
• This approach offers a potentially more efficient alternative to genome-wide epigenotyping for age estimation.