Explainable artificial intelligence identifies an AQP4 polymorphism-based risk score associated with brain amyloid...

Area of Science:

• Neuroscience
• Genetics
• Biochemistry

Background:

• The glymphatic system clears brain waste, including amyloid-beta (Aβ).
• Aquaporin-4 (AQP4) is a key protein in the glymphatic system, and its genetic variations may affect Aβ clearance and Alzheimer's disease (AD) progression.
• Previous studies suggest AQP4 genetic variations impact AD clinical outcomes and sleep patterns.

Purpose of the Study:

• To investigate the relationship between a risk score derived from AQP4 single-nucleotide polymorphisms (SNPs) and Aβ neuropathology in cognitively unimpaired older adults.
• To explore synergistic effects of AQP4 SNPs on brain amyloid burden using machine learning and explainable AI.

Main Methods:

• Utilized data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort for machine learning analysis.
• Developed a sex-specific AQP4 SNP-based risk score.
• Validated the risk score using data from the screening phase of the A4 study.

Main Results:

• A significant association was found between the AQP4 SNP-based risk score and brain amyloid burden in both the ADNI and A4 study cohorts.
• The findings indicate that multiple AQP4 SNPs may act synergistically to influence the accumulation of brain amyloid.

Conclusions:

• The results support the hypothesis that the glymphatic system, specifically AQP4, plays a role in brain amyloid aggregation pathology.
• The study suggests that AQP4 genetic variations are linked to amyloid burden, potentially impacting Alzheimer's disease risk.