Transcriptome analysis of Alzheimer's disease identifies links to cardiovascular disease

Understanding the pathogenesis in the early stages of late-onset Alzheimer’s disease (AD) can help in gaining important mechanistic insights into this devastating neurodegenerative disorder. Integration of multiple computational approaches to address the different levels of information embedded in microarray data, such as networks of coexpressed genes, functional annotation modules, and cis-regulatory elements shared by 1 co-expressed genes, leads to greater understanding of complex diseases such as AD. We use our recently developed methods for co-expression network analysis (CoExp) and genome-wide motif identification (WordSpy) along with functional annotation clustering on single cell expression data to analyse AD. CoExp automatically identified 6 clusters/modules, each of which represented a biological process perturbed in AD. Interestingly, AD related genes like APOE, A2M, PON2, MAP4 and cardiovascular diseases (CVD) associated genes such as COMT, CBS, WNK1 all congregated in one of the six modules. This module that contained 18 disease associated (cardiovascular, neurodegeneration, diabetes, stroke) genes had the maximum number of hub genes. Some of the disease related genes were also hub genes while many of them were directly connected to one or more hub genes. Further investigation of this disease associated module unveiled significant cis-regulatory elements that were significantly similar to the binding sites of transcription factors involved in AD and CVD. Our results showcase extensive links between genes associated with AD and CVD at the co-expression and co-regulation levels and provide strong supporting evidence to the hypotheses linking CVD and AD.