Predicting progression of Alzheimer's disease using blood-based multi-omics data

Area of Science:

• Neuroscience
• Genetics
• Biomarker Discovery

Background:

• Early prediction of Alzheimer's disease (AD) progression from mild cognitive impairment (MCI) is crucial for timely intervention.
• Non-invasive biomarkers for distinguishing progressive MCI (pMCI) from stable MCI (sMCI) are needed.
• Identifying individuals at risk of converting to AD can significantly improve clinical management.

Purpose of the Study:

• To develop a machine learning (ML) framework for early prediction of AD progression.
• To integrate blood-based multi-omics and demographic data to differentiate pMCI from sMCI.
• To identify reliable, interpretable, blood-based biomarkers for AD progression.

Main Methods:

• Trained ML models using combinations of single nucleotide polymorphism (SNP), DNA methylation, gene expression, lipid, and bile acid metabolite data.
• Employed early and late data integration strategies, with late integration showing superior performance.
• Utilized LIME and SHAP for feature interpretability to identify key predictive biomarkers.

Main Results:

• Late data integration consistently outperformed early integration in predicting MCI progression.
• The L1-regularized logistic regression model achieved a high F1 score of 90.7% when combining SNP and lipid data.
• Identified reproducible biomarkers including specific SNPs, methylation changes, and lipid levels associated with AD pathology.

Conclusions:

• Combining multi-omics and demographic data enhances the early prediction of AD progression from MCI.
• Blood-based, interpretable biomarkers are feasible for precision diagnostics in AD.
• The developed ML framework shows promise for improving early detection and management of Alzheimer's disease.