Research Project
10300802
Decoding the Impact of Sex Differences on Alzheimer?s Disease Risk The molecular basis and genetic architecture of Alzheimer?s Disease (AD) remain poorly defined. Solving these problems is further complicated by the differences that exist between men and women with respect to the prevalence, onset, progression and comorbidities of AD, suggesting that some contributing genetic variants are sex-specific. So far, mixed-gender Genome-wide association studies (GWAS) have linked over 100 loci with AD. These loci explain much of the population-attributable risk but just a fraction of heritability, and with no distinction between men and women. It is unlikely that this heritability gap would improve just by splitting studies by sex, however, since separate analyses on about half as many patients would be less powerful. Rather, in order to design effective surveillance, screening, preventive, and stratification programs tailored to each sex, there is a critical need for more sensitive and accurate methods, able to compute the link between genetic variants and AD risk separately and specifically in men and women. To do so, we propose an integrative computational approach that will be validated by experimental and translational studies of candidate genes. Rather than seek individual variants, we focus instead on genes and their coding regions. In order to increase the power of our studies, we developed an unbiased evolution-based continuous score for the functional impact of any coding variant, from 0 (neutral) to 1 (complete loss of function). Using this scoring system adds to the usual analyses of human variants a vast number of amino acid mutation experiments already performed by evolution over billions of years, with each mutation being tied to a functional readout based on the context of its phylogenetic divergence. With this score, we now propose to identify genes that carry significantly more impactful coding variants in AD women, or AD men, compared to sex-matched controls. The gain in statistical power of this approach compared to GWAS has been demonstrated in preliminary data. In Aim 1 we propose to discover sex-specific AD genes on more than 1000 Alzheimer?s Disease Sequencing Project (ADSP) men and 1400 ADSP women; and in Aim 2 to discover sex-specific modifiers of APOE. Aim 3 will include validation of computationally-derived candidate genes in human brain tissue and cerebrospinal fluid (CSF), and thorough experimental characterization in AD animal-models (mouse and Drosophila). Together, this combination of novel, integrative computational approaches and multi-model systems validation experiments will yield new biomarkers that improve sex-specific risk stratification of AD status and reveal differences in disease mechanisms between women and men that highlight potential therapeutic targets specific to each.
