SUMO-Related Gene Biomarkers in Alzheimer's Disease via Bioinformatics Analysis

Bioinformatics analysis of Alzheimer's disease data intersecting 189 small ubiquitin-related modifier-related genes with 12,853 differentially expressed genes. Jing Xu published this research dataset on figshare in March 2026. The dataset supports the identification of candidate biomarkers like CREBBP, PIAS1, and TRIM28.

Use Cases

• Validate biomarker candidates like CREBBP and TRIM28 using ROC curve analysis on gene expression data.
• Perform gene set enrichment analysis (GSEA) to link identified feature genes to pathways such as olfactory transduction and lysosome function.
• Construct a protein-protein interaction network to identify top 15 core genes from candidate gene lists.
• Apply SVM-RFE algorithm to select feature genes from the 107 candidate genes identified.
• Analyze immune infiltration patterns using gene expression data to explore immune cell involvement in AD progression.

Strengths

• Analysis integrates data from three public gene expression datasets (GSE140831, GSE63060, GSE140830).
• Identifies 107 candidate genes from intersecting 189 SRGs with 12,853 DEGs.
• Validation performed via RT-qPCR confirming expression trends for CREBBP and TRIM28.

Limitations

• Dataset size is 72.0 KB, indicating limited raw data scope.
• Underlying sample sizes and specific column structures for the gene expression datasets are not provided.
• Biomarker validation is based on specific datasets and may require replication in independent cohorts.

Provenance

Source

Public gene expression databases (GEO).

Collection Method

Bioinformatics analysis intersecting DEGs with SRGs, followed by PPI network construction and SVM-RFE modeling.

Freshness

Last updated March 2026.