A Network‐based Deep Learning Framework Translates GWAS and Multi‐Omics Findings to Pathobiology and Drug Repurposing for Alzheimer's Disease. (20th December 2022)

• Record Type: Journal Article
• Authors: Xu, Jielin; Hou, Yuan; Zhou, Yadi; Bekris, Lynn M.; Pieper, Andrew A.; Cummings, Jeffrey; Leverenz, James B.; Cheng, Feixiong
• Is Part Of: Alzheimer's & dementia. Volume 18(2022)Supplement 4
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-20
• 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.066647
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Human genome sequencing studies have identified numerous loci associated with complex diseases, including Alzheimer's disease (AD). However, translating human genetic and genomic findings (i.e., genome‐wide association studies [GWAS]) to pathobiology and therapeutic discovery remains a major challenge. Today, artificial intelligence and deep learning approaches that identify new risk genes and drug targets from human genome sequencing findings are enabling a more complete mechanistic understanding of disease biology. This is facilitating more rapid development of targeted therapeutic interventions for AD. Method: We presented a net work t opology‐based deep learning framework to identify disease‐a ssociated g enes (NETTAG). NETTAG integrates multi‐genomics data and the human protein‐protein interactome to infer putative risk genes and drug targets impacted by GWAS loci. Specifically, we leveraged non‐coding GWAS loci effects on expression quantitative trait loci (eQTLs), histone‐QTLs, transcription factor binding‐QTLs, enhancers and CpG islands, promoter regions, open chromatin, and promoter flanking regions. The fundamental premise of NETTAG is that disease risk genes exhibit distinct functional characteristics compared to non‐risk genes, and can therefore be distinguished by their aggregated genomic features under the human protein interactome. Result: Applying NETTAG to the latest AD GWAS data, we identified 156 putative AD‐risk genes (i.e., APOE, BIN1, GSK3B, MARK4, and PICALM ). We showed that predicted risk genes are: 1) significantly enriched in AD‐related pathobiological pathways, 2) more likely to be differentially expressed in transcriptomes and proteomes of AD brains, and 3) enriched in druggable targets with approved medicines. Specifically, we showed that NETTAG‐predicted genes (e.g., MEF2D, CPLX2, KLF4, ACTL6B, P2RX7 and etc.) are differentially expressed in AD‐associated microglia and astrocytes from single‐nuclei RNA‐sequencing data of human postmortem brains with varying degrees of AD neurobiology. Via network‐based prediction, we identified multiple repurposable drug candidates (i.e., choline, deferoxamine and ibudilast) for potential treatment of AD. Conclusion: Our findings suggest that understanding human pathobiology and therapeutic development could benefit from network‐based deep learning methodology that utilizes GWAS findings with multimodal genomic analyses.