Premium
Protein‐protein interaction networks of genes associated with different cognitively defined subtypes of late‐onset Alzheimer's disease in five white populations identify novel candidate genes
Author(s) -
Mukherjee Shubhabrata,
Mez Jesse,
Trittschuh Emily H.,
Saykin Andrew J.,
Gibbons Laura E.,
Sanders R. Elizabeth,
Fardo David W.,
Crane Paul K.
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.045014
Subject(s) - genome wide association study , genetic association , candidate gene , biology , interactome , dementia , genetics , population , gene , disease , clinical dementia rating , computational biology , medicine , single nucleotide polymorphism , genotype , environmental health
Background We previously defined late‐onset Alzheimer's disease (AD) subgroups using cognitive profiles (memory, executive functioning, language, and visuospatial ability) for individuals with Clinical Dementia Rating < 2 in five White populations. This approach results in six subgroups: those with no domain with a substantial relative impairment; those with an isolated substantial relative impairment in one of four domains, and those with multiple domains (mixed) with substantial relative impairments. We performed gene network analysis to find unique and overlapping genes to enhance our understanding of AD heterogeneity. Method We performed genome‐wide association studies (GWASs) for each of the AD subtypes against cognitively normal elderly adjusting for age, sex, and population substructure using PLINK v1.9 in each population and meta‐analyzed results using METAL. We excluded the AD‐Executive function group because of small sample size. For each of the subtypes, we performed gene‐wide association analyses using VEGAS v2 and combined results with human protein‐protein interaction (PPI) data using a dense module searching method to identify candidate gene modules. This approach searches the entire interactome and exhaustively evaluates the combined effects of multiple genes to identify networks of interacting genes with low p ‐values for the phenotype of interest. We performed gene network analysis for AD using Stage 1 GWAS summary statistics reported in the 2019 IGAP GWAS of AD paper and compared genes in the top module for AD to the top module for each AD subtype. Result The top gene network module for AD and AD subtypes were enriched for genes in Chromosome 19. Ubiquitin C ( UBC ) was the hub gene in each of the top modules (See Figure 1). Very few genes were common across the different subtypes and with AD, and except for UBC , all of them reside in Chromosome 19 (See Figure 2). Conclusion Network analyses using PPI data identified novel candidate genes for cognitively‐defined AD subtypes, which supports the AD heterogeneity idea. Further functional enrichment analysis is needed to determine whether these candidate loci may provide targets for interventions to ameliorate AD.