Alzheimer's disease (AD) can be divided into two categories based on age of onset and familial aggregation, familial (FAD), and late onset (LOAD). While much is known about genetic risk factors for FAD, the vast majority of AD cases are LOAD. Our knowledge of the mechanisms by which the known genetic risk factors for AD alter risk clearly indicates that amyloid metabolism and deposition is central to the pathology of AD. To date, studies using the endophenotype-based approach have been biased toward the study of existing candidate genes. This can be easily addressed by the performance of genome-wide association studies (GWAS) in samples with endophenotype information. The ability to measure differences in total expression, splicing, or transcript ratios between AD cases and controls will further inform our future studies, making it possible to identify genetic variants which have direct effects on expression.