Admixture mapping identifies novel regions influencing Alzheimer disease in African Americans. (1st February 2022)

• Record Type: Journal Article
• Authors: Hamilton‐Nelson, Kara L.; Rajabli, Farid; Kunkle, Brian W.; Tosto, Giuseppe; Reitz, Christiane; Naj, Adam C; Whitehead, Patrice L.; Kushch, Nicholas A.; Beecham, Gary W.; Byrd, Goldie S.; Bush, William S.; Mayeux, Richard; Farrer, Lindsay A.; Haines, Jonathan L.; Schellenberg, Gerard D.; Pericak‐Vance, Margaret A.; Martin, Eden R
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 3
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-02-01
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.056443
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: African Americans (AA) are substantially underrepresented in Alzheimer's disease (AD) genetic studies, yet their admixed genetic ancestry (African and European) provides a unique opportunity to identify novel genetic factors associated with AD related to genetic ancestry. Admixture mapping (AM) provides a more powerful approach than SNP‐based genome‐wide association studies (GWAS) in admixed populations in part due to the lower multiple testing burden. In this study we used AM to identify regions associated with AD in AA individuals. Methods: Our analyses included 10, 271 individuals from 17 AD Sequencing Project and AD Genetics Consortium cohorts. We estimated global ancestry (GA) using the GENESIS software. To infer local ancestry (LA), the target AA dataset was combined with appropriate reference‐population samples from HGDP reference panel, and LA was estimated using SHAPEIT followed by RFMix. Then, we performed AM using the GENESIS software separately on each cohort. We meta‐analyzed the AM results with the random effect approach (RE2). We calculated the significance threshold using STEAM software. Finally, we performed logistic regression of genotype on affection status for variants across the prioritized regions from AM for fine‐mapping. The regression model included LA and genotype as main effects and term for their interaction, along with GA, sex and age as covariates, and used permutation‐based testing approach for multiple test correction (N=10, 000). Results: AM identified two genome‐wide significant loci on chromosomes 17p13.2 (pv=2.2 x10 ‐5 ) and 18q21.33 (pv=1.22x10 ‐5 ). 17p13.2 region was identified as a genome‐wide significant in two previous studies in non‐Hispanic White population. To fine map this region we conducted ancestry‐aware regression analysis. LA x genotype interaction model found the MINK1 gene (rs72835013) on the 17p13.2 region significantly associated with AD (pv = 1x10 ‐4 ). Conclusions: Our results confirmed an AD associated region on the chromosome 17p13.2 and showed that the 17p13.2 region increases AD risk in AA individuals with the European LA. This region includes genes previously indicated in AD such as SCIMP through association studies and the MINK1 and SLC25A11 genes through brain expression studies.