Biomarkers.

Area of Science:

• Biochemistry
• Gerontology
• Proteomics

Background:

• Human aging and disease are increasingly modeled using multi-organ frameworks.
• Plasma proteomics is a key tool for predicting biological age, yielding the proteome-based biological age gap (ProtBAG).
• This study utilizes cerebrospinal fluid (CSF) proteomics to expand organ-specific aging clocks.

Purpose of the Study:

• To derive 11 organ-specific proteome-based biological age gaps (ProtBAGs) using CSF proteomics data.
• To apply two distinct machine learning methods for developing these organ-specific aging clocks.
• To evaluate the performance of ProtBAGs across different organ systems.

Main Methods:

• CSF proteomics data from the ADNI study (7,008 proteins, 736 participants) were analyzed.
• Organ-enriched proteins were identified, and two machine learning models (Linear SVR and LASSO) were trained.
• Model performance was assessed using mean absolute error (MAE) and Pearson's correlation coefficient (r) via cross-validation.

Main Results:

• Proteomics imputation with a 2% missing rate yielded the best model performance (r²=0.54).
• Brain and hepatic ProtBAGs demonstrated the lowest mean absolute error (MAE) values, indicating higher accuracy.
• Brain and hepatic ProtBAGs also exhibited the highest Pearson's r values, confirming robust predictive power.

Conclusions:

• 11 organ-specific ProtBAGs were successfully developed using CSF proteomics from the ADNI cohort.
• This work enhances the existing organ aging clock framework with CSF-based multi-organ insights.
• Future studies will explore the links between these novel ProtBAGs, cognitive function, and Alzheimer's disease progression.