A multiple-causal-gene-cluster model underlying GWAS signals of Alzheimer’s disease

Genome-wide association studies (GWASs) have identified dozens of genetic susceptibility loci for Alzheimer’s disease (AD). Nevertheless, the underlying causal variants and biological mechanisms remain elusive. Here, we systematically integrated AD GWAS with comprehensive multi-omics data, and distilled 304 potentially functional variants and 166 causal genes from 49 loci. Intriguingly, we found that most of AD GWAS loci contain multiple functional variants and causal genes. In vitro assays showed that one functional variant regulated multiple genes in the 11p11.2 locus (the CELF1/SPI1 locus) and alteration of these target genes contributed to AD-related molecular processes, supporting the co-existence of multiple functional variants and AD-relevant causal genes within a single locus. We thus proposed a multiple-causal-gene-cluster model that co-dysregulation of a cluster of genes within a single GWAS loci individually or synergistically contribute to AD development. This model provides a novel insight into the biological mechanisms underlying the GWAS loci of complex traits.