Dept. of Medical Informatics and Clinical Epidemiology
Oregon Health & Science University
Alzheimerâs Disease (AD) is the leading cause of dementia in the United States yet the genetics behind this complex disease remains unclear. With the exception of Apolipoprotein E e4 (APOE-e4), more than 40 loci have been implicated as common genetic risk factors of AD but none of these have been confirmed. We completed a genome wide association study on 567 unrelated participants in the Alzheimerâs Disease Neuroimaging Initiative (ADNI) data set. DNA samples were genotyped with the Illumina Human610 Quad BeadChip and 543,715 single nucleotide polymorphisms (SNPs) were included after undergoing quality control measures. Genome wide association studies (GWASs) have successfully identified genetic associations to individual phenotypes in a univariate framework across many complex diseases including AD. However, the effort to detect pleiotropic associations, where multiple traits are associated with the same genetic loci, is far less common and has never been tried in an AD GWAS. Two multivariate methods, Principle Components Analysis (PCA) and Seemingly Unrelated Regression (SUR), were employed to determine the genetic 8 association of three quantitative correlated endophenotypes of AD. The PCA method incorporated hippocampal volume, ventricular volume and cognitive memory tests and the SUR method included hippocampal volume and cognitive memory tests. Our study identified 23 unique SNPs, with six SNPs found in common between both methods after adjusting for any biases. The PCA method found 21 SNPs (p-value < 10[superscript -5]) and the SUR method found eight SNPs (p-value < 10[superscript -5]). All of the identified genes have not been otherwise linked to AD, indicating a multivariate framework can provide new insight to genetic research of these phenotypes and AD.
School of Medicine
Bhatt, Priya, "Assessing multivariate analysis of GWAS for identification of genetic variants in Alzheimer's disease" (2012). Scholar Archive. 868.