Protein‐protein interaction networks reveal paths connecting functional domains associated in Alzheimer's disease. (December 2021)

• Record Type: Journal Article
• Authors: Gockley, Jake; Cary, Gregory A.; Wiley, Jesse C; Poehlman, William L.; Heath, Laura M; Leal, Karina; Greenwood, Anna K; Carter, Gregory W; Mangravite, Lara M
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 3
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12
• 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.057673
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Alzheimer's Disease (AD) is an incurable neurodegenerative disease comprising 60‐70% of dementia diagnoses. Despite known genetic risk factors and pathology of disease, AD mechanisms remain to be elucidated. To identify therapeutic targets in the context of AD mechanisms, we constructed a multi‐modal neocortical protein‐protein interaction network with integrated AD risk scores to track AD disease through protein interactions. Method: The Pathway Commons database was filtered for genes expressed in 9 brain regions from the Accelerated Medicines Partnership in AD (AMP‐AD). This network of 13, 867 genes and 806, 950 interactions was integrated with co‐expression networks yielding topology of interaction communities. Gene weights were derived from the genetic, transcriptome (RNA‐Seq), and proteome (LFQ and TMT) components of the Target Enablement to Accelerate Therapy Development for Alzheimer's Disease (TREAT‐AD) score, a score of overall AD‐Risk for every gene compiled across ten components. We wrote an R package to trace paths enriched in AD risk between input target and sentinel genes. We identified 15 AD associated biological domains (domains) containing GO Terms enriched in high risk AD genes. Genes from these domains were filtered by TREAT‐AD scores into the top AD‐Candidates by biodomain. Each list of genes within a biodomain was traced pairwise to the genes within the 15 lists creating an AD subnetwork of 3, 015 genes and 87, 615 protein interactions. Result: This subnetwork represents a valuable resource to explore mechanistic paths of AD biology with respect to their contextual environments. Biodomains most enriched for AD‐risk after tracing were Mitochondria and Metabolism, Synaptic Dysfunction, Structural Stabilization, Endolysosomal Dysfunction, and Immune Response. Conclusion: We created a valuable resource and framework to extract AD‐relevant pathways from a massive cortical network of protein‐protein interactions. Our subnetworks identify paths with AD mechanisms that connect relevant domains driving AD. This resource provides needed context for the community to use when designing therapeutic target candidates or hypotheses relevant to AD disease mechanisms. We hope that elucidating the relevant pathways contributing to AD disease progression, along with their contextual interactions, inform the communities' priors when designing experiments to alter AD pathways within model systems.