Co-expression Network Analysis Reveals Novel Genes Underlying Alzheimer’s Disease Pathogenesis

Abstract

Background: The pathogenesis of Alzheimer’s disease (AD) remains to be elucidated. This study aimed to identify the hub genes in AD pathogenesis and determine their functions and pathways.Methods: A co-expression network for an AD gene dataset with 401 samples was constructed, and the AD status-related genes were screened. The hub genes of the network were identified and validated by an independent cohort. The functional pathways of hub genes were analyzed.Results: The co-expression network revealed a module that related to the AD status, and 101 status-related genes were screened from the trait-related module. Gene enrichment analysis indicated that these status-related genes are involved in synaptic processes and pathways. Four hub genes (ENO2, ELAVL4, SNAP91, and NEFM) were identified from the module, and these hub genes all participated in AD-related pathways, but the associations of each gene with clinical features were variable. An independent dataset confirmed the different expression of hub genes between AD and controls.Conclusions: Four novel genes associated with AD pathogenesis were identified and validated, which provided novel therapeutic targets for AD.

Preservation Analysis on Spatiotemporal Specific Co-expression Networks Suggests the Immunopathogenesis of Alzheimer’s Disease

“…In the current study, we explored AD-related biological processes by analysing the coexpression gene modules of different brain regions in different disease stages. The coexpression network features and key genes of AD peripheral blood or brain have been reported in several previous studies (Seyfried et al, 2017;Liang et al, 2018;Sweeney et al, 2018;Zhang et al, 2018;Hu et al, 2020;Kelly et al, 2020;Soleimani Zakeri et al, 2020). Among these studies, Wang et al performed a pan-cortical brain region genomic analysis, obtained and ranked 44,692 gene probesets, 1,558 coexpressed gene modules and 19 brain regions based upon their association with AD; through these analyses temporal lobe gyri were identified as sites associated with the greatest and earliest gene expression abnormalities, abnormal expression was specific to cell type of oligodendrocytes, astrocytes, and neurons, and neurobiological pathways (included actin cytoskeleton, axon guidance, and nervous system development) were enriched by abnormally expressed genes and modules (Wang et al, 2016); however, the changes in coexpression modules in sub-brain regions during AD development have not been fully studied.…”

“…In addition to finding differentially expressed genes (DEGs) that are significantly changed in AD patients, expression profiling can also provide more evidence about the systematic molecular processes underlying the etio-pathogenesis of AD. Based upon the associations between coexpressed gene modules and AD traits, several previous studies identified AD-related gene modules, which suggests that the biological processes that these genes contribute to may be affected in AD (Wang et al, 2016;Tang and Liu, 2019;Hu et al, 2020;Kelly et al, 2020). By using spatial-temporal expression pattern analysis, transcriptome data can also provide evidence about specific brain regions and cell types that are possibly related to AD (Wang et al, 2016).…”

A Novel 16-Genes Signature Scoring System as Prognostic Model to Evaluate Survival Risk in Patients with Glioblastoma

“…It also has been found can affect many cellular processes, such as homeostasis, learning and memory, cancer, and pain [ 50 ]. SNAP91 is associated with Alzheimer’s disease [ 51 ], schizophrenia [ 52 ], Parkinson’s disease [ 53 ] and colorectal cancer [ 54 ]. PCSK1N has an association with various neurodegenerative diseases, widely expressed in neurons throughout the brain [ 55 , 56 ].…”

Deep Learning-based Feature Extraction with MRI Data in Neuroimaging Genetics for Alzheimer’s Disease

“…One of the reasons may be that apart from the major effect from one brain ROI, the principal components may have information from other brain regions as well. For instance, the 3rd PC of the augmented model was found associated with rs 1158059 from gene SNAP91 which has also been found to be involved in AD pathways [41]; the 7th principal component of the multi-branch model with the whole brain method identifies SNP rs 10514441 related to gene WWOX which plays an important role AD through interactions with its protein partners and cell pathology and degeneration [39]; These brain regions [33, 16] have been previously shown to be associated with AD. Moreover, the SNP rs 2075650 associated with the 2nd principal component (Table 1) has been shown to be strongly associated with hippocampus volume of the brain [34, 29].…”