As genome-wide association studies (GWAS) have grown in size, the number of genetic variants that have been associated per disease has correspondingly increased. Despite this increase in the number of single-nucleotide polymorphisms (SNPs) identified per disease, their biological interpretation has in many cases remained elusive. To address this, we have combined GWAS results with orthogonal sources of evidence, namely the current knowledge of molecular pathways; real-world clinical data from six ...
As genome-wide association studies (GWAS) have grown in size, the number of genetic variants that have been associated per disease has correspondingly increased. Despite this increase in the number of single-nucleotide polymorphisms (SNPs) identified per disease, their biological interpretation has in many cases remained elusive. To address this, we have combined GWAS results with orthogonal sources of evidence, namely the current knowledge of molecular pathways; real-world clinical data from six million patients; RNA expression across tissues from Alzheimer's disease (AD) patients, and purpose-built rodent models for experimental validation. In more detail, first we show that when examined at a pathway level, analysis of all GWAS studies groups AD in a cluster with disorders of immunity and inflammation. Using clinical data, we show that the degree of comorbidity of these diseases with AD correlates with the strength of their genetic association with molecular participants in the Janus kinases/signal transducer and activator of transcription (JAK-STAT) pathway. Using four independent RNA expression datasets we then find evidence for the altered regulation of JAK-STAT pathway genes in AD. Finally, we use both in vitro and in vivo rodent models to demonstrate that Aβ induces gene expression of the key drivers of this pathway, providing experimental evidence to validate these data-driven observations. These results therefore nominate JAK-STAT anomalies as a prominent aetiopathological event in AD and hence a potential target for therapeutic development, and moreover demonstrate a de novo multi-modal approach to derive information from rapidly increasing genomic datasets.
+
