Fig. 1.

Overview of research strategy. The strategy is based on the hypothesis that there may be variants that affect LOAD risk by influencing gene expression, and such variants would be associated with both AD risk and gene expression levels. We systematically integrated LOAD GWAS and eQTL data with the Sherlock algorithm. The top signals identified by Sherlock were then replicated in independent LOAD and eQTL data sets. We also tested whether the positive risk SNP, rs2927438, modified LOAD risk independent of APOE ε4 status. Finally, we explored its association with several endophenotypes, including AAO of LOAD, hippocampal volume, and cognitive performance. Abbreviations: AAO, age at onset; AD, Alzheimer’s disease; eQTL, expression quantitative trait loci; LOAD, late-onset Alzheimer’s disease; GWAS, genome-wide association study; SNP, single-nucleotide polymorphism.