Integrative Gene Co-expression Network Analysis Reveals Protein-Coding and lncRNA Genes Associated with Alzheimer's Disease Pathology
IntroductionAlzheimers disease (AD) is a complex neurodegenerative disorder involving widespread molecular disruptions, many of which remain poorly understood. While AD pathology progresses through distinct brain regions, it is unclear whether these regions are similarly affected. Long non-coding RNAs (lncRNAs) have emerged as key regulators in cellular processes, but their roles in AD remain largely unexplored. A comprehensive analysis of the complex molecular interactions underlying AD, particularly the roles of lncRNAs in their interactions with protein-coding genes and the comparison of networks across brain regions, could offer valuable insights into the diseases progression and underlying mechanisms.

MethodTo address this, we applied consensus weighted gene co-expression network analysis using a meta-analytic approach to 396 postmortem brain RNA-seq samples to explore AD pathophysiology mechanisms.

ResultsOur study revealed significant network rewiring in AD, with pronounced alterations in the temporal cortex compared to the frontal cortex. While the temporal cortex showed adaptive changes in gene interactions, the frontal cortex exhibited a loss of healthy correlations, potentially reflecting different levels of disease progression. We identified key players in the temporal cortex AD network, including 46 protein-coding genes and 27 lncRNAs. Using the established functions of protein-coding genes as anchors, we provided functional annotations for over 100 lncRNAs across two brain regions, identifying potential lncRNAs involved in AD pathology and highlighting their roles in both healthy and diseased states.

ConclusionWe provided novel insights into the molecular interactions underlying AD and introduced new candidate protein-coding and lncRNA genes for further experimental validation and therapeutic exploration.