Identification of key candidate genes for IgA nephropathy using machine learning and statistics based bioinformatics...

Area of Science:

• Nephrology
• Bioinformatics
• Genomics

Background:

• Immunoglobulin-A nephropathy (IgAN) is a kidney disease characterized by IgAN deposits, inflammation, and tissue damage.
• Current understanding of IgAN's molecular mechanisms and progression requires further exploration.
• Bioinformatics approaches are crucial for identifying novel candidate genes and pathways in IgAN.

Purpose of the Study:

• To identify key candidate genes for IgAN using integrated machine learning and statistical bioinformatics models.
• To elucidate molecular mechanisms underlying IgAN development and progression.

Main Methods:

• Differential gene expression analysis using limma.
• Gene enrichment analysis via DAVID.
• Protein-protein interaction network construction (STRING, Cytoscape) to identify hub genes and modules (MCODE).
• Machine learning algorithms (SVM, LASSO, PLS-DA) for discriminative gene identification.

Main Results:

• Identification of differentially expressed genes (DEGs) in IgAN.
• Construction of a protein-protein interaction network revealing key hub genes and modules.
• Selection of discriminative genes using SVM, LASSO, and PLS-DA.
• Identification of five overlapping key candidate genes: FOS, JUN, EGR1, FOSB, and DUSP1.

Conclusions:

• FOS, JUN, EGR1, FOSB, and DUSP1 are identified as critical candidate genes for IgAN.
• These genes hold potential for improving IgAN diagnosis and therapeutic strategies.
• The study provides a foundation for further research into IgAN pathogenesis.