Background: Alzheimer's disease (AD) is an irreversible neurodegenerative disorder with no fully curative treatment.
Aim: This study aims to identify effective biomarkers for AD diagnosis and treatment by combining multi-omics and Mendelian randomisation (MR) analyses.
Subjects and methods: Positron emission tomography (PET), single nucleotide polymorphism (SNP), and gene expression data of AD patients using advanced correlation analysis methods (AdaSMCCA, rAdaSMCCA, and unAdaSMCCA algorithms) are integrated.
Results: Several regions of interest, risk SNP sites, and risk genes associated with AD are identified. Expression quantitative trait loci (eQTL) for the top 100 risk genes are retrieved from public datasets. A two-sample MR analysis using genome-wide association study (GWAS) data reveals two genes (FAM117A and ACSL1) causally related to AD. Additionally, single-cell transcriptome (scRNA-seq) data from AD samples are analysed to identify high-scoring cell clusters and their interactions.
Conclusions: The identified multi-omics biomarkers and genes causally related to AD could inform clinical diagnosis and treatment.