Weighted burden analysis of exome‐sequenced late‐onset Alzheimer's cases and controls provides further evidence for a role for PSEN1 and suggests involvement of the PI3K/Akt/GSK‐3β and WNT signalling pathways

Abstract

Previous studies have implicated common and rare genetic variants as risk factors for late‐onset Alzheimer's disease (LOAD). Here, weighted burden analysis was applied to over 10,000 exome‐sequenced subjects from the Alzheimer's Disease Sequencing Project. Analyses were carried out to investigate whether rare variants predicted to have a functional effect within a gene were more commonly seen in cases or in controls. Confirmatory results were obtained for TREM2, ABCA7, and SORL1. Additional support was provided for PSEN1 (p = 0.0002), which previously had been only weakly implicated in LOAD. There was suggestive evidence that functional variants in PIK3R1, WNT7A, C1R, and EXOC5 might increase risk and that variants in TIAF1 and/or NDRG2 might have a protective effect. Overall, there was strong evidence (p = 5 × 10−6) that variants in tyrosine phosphatase genes reduce the risk of developing LOAD. Because PIK3R1 variants are expected to impair PI3K/Akt/GSK‐3β signalling while variants in tyrosine phosphatase genes would enhance it, these findings are in line with those from animal models, suggesting that this pathway is protective against Alzheimer's disease.

Multiple Linear Regression Allows Weighted Burden Analysis of Rare Coding Variants in an Ethnically Heterogeneous Population

“…The full set of weights is shown in online supplementary Table 1 (for all online suppl. material, see www.karger.com/doi/10.1159/000512576), copied from the previous reports which used this method [6, 7]. Variants were excluded if there were >10% of genotypes missing or if the heterozygote count was smaller than both homozygote counts.…”

“…The comparison of variant burden was then implemented in a ridge logistic regression framework and this allowed the inclusion of covariates, such as population principal components, as well as additional risk factors, such as pathogenic copy number variants and polygenic risk scores [5]. These approaches were applied to large samples of exome sequenced cases and controls and implicated genes affecting functioning of the glutamatergic NMDA receptor in schizophrenia and genes coding for tyrosine phosphatases in late-onset Alzheimer’s disease [6, 7].…”

Dissecting the clinical heterogeneity of early-onset Alzheimer’s disease

“…Analysis of LOAD families has more recently resulted in the identification of rare, risk-conferring variants and established pathogenic variants in APP , PSEN1 , and PSEN2 , thus suggesting that these genes may also be relevant for the more common, late-onset variety of AD [ 162 , 163 ]. Additional studies using data from the AD Sequencing Project also provide suggestive evidence that rare variation in PSEN1 increases risk for LOAD [ 164 , 165 ]. Given the potent influence that the amyloid cascade hypothesis [ 166 , 167 ] has had on the field and on AD drug development, the finding that variants in these genes may also confer risk for LOAD provides important support for the generalizability of this hypothesis (now updated [ 168 , 169 ]) to all forms of AD.…”

Genetic and Multi-omic Risk Assessment of Alzheimer’s Disease Implicates Core Associated Biological Domains

“…Genetic Considerations in Selecting Biological Domains. In genetics, we focused on key genome wide association studies (GWAS), the newer genome wide association study by proxy (GWAX) [15][16][17][18][19][74][75][76][77][78][79][80] using the parental disease status, and whole exome sequencing studies [81][82][83][84][85][86][87][88][89][90][91][92][93][94][95] that have transpired over the last decade. The identification of potential genetic risk associated with individual genes is represented in the genetics score (detailed below), and the goal here is not to recapitulate the scoring methodology, but to assess the potential biological contexts of the imputed genes' biological function.…”