1. Identification of DEGs
We identified a total of 96 samples, comprising 47 IgAN samples and 49 normal samples, in the two datasets. Based on the cutoff criteria of |log2 FC| ≥1.0 and adj Pvalue ≤0.05, 148 overlapping DEGs were showed by a Venn diagram obtained from IgAN group vs. control group, consisted of 53 upregulated and 95 downregulated. The results on the expression level analysis are presented in a volcano plot in Fig. 1A. As indicated in the clustering heat map (Fig. 1B), these DEGs could well distinguish the IgAN and control group completely.
2.Gene Ontology and KEGG Analysis of DEGs
To investigate the biological classification of DEGs, the overall genes in three ontologies were identified using DAVID. The cut-off criterion was set as P < 0.05. The GO function annotation is divided into three functional groups, cell component (CC), molecular function (MF), and biological process (BP). The CC terms of the DEGs were significantly enriched in extracellular exosome, region and space (Fig.2A,3A). The MF terms were mainly enriched in transcriptional activator activity and RNA polymerase II core promoter proximal region sequence-specific binding; heme binding identical protein binding (Fig.2B,3B). The changes in BP were significantly enriched in response to cAMP., cellular response to fibroblast growth factor stimulus and inflammatory response (Fig2C,3C).
The KEGG pathway enrichment analysis revealed that the DEGs were mainly enriched in Protein digestion and absorption, Pertussis and Staphylococcus aureus infection (Fig.2D,3D).
3. Construction of the PPI network and module analysis
To further investigate the interaction among the 148 DEGs, a PPI network was constructed from STRING(Fig.4A) The most significant module was obtained using Cytoscape MCODE plug-in. The module consisted of 15 nodes and 89 edges, include Fc fragment of IgE receptor Ig (FCER1G), HCK proto-oncogene, Src family tyrosine kinase(HCK), TYRO protein tyrosine kinase binding protein(TYROBP),V-set and immunoglobulin domain containing 4(VSIG4),colony stimulating factor 1 receptor(CSF1R),complement C1q A chain(C1QA), complement C3a receptor 1(C3AR1),cytochrome b–245 beta chain(CYBB), hematopoietic cell-specific Lyn substrate 1(HCLS1), integrin subunit beta 2(ITGB2), interleukin 10 receptor subunit alpha (IL10RA), lysosomal protein transmembrane 5(LAPTM5), neutrophil cytosolic factor 2(NCF2),CD48 and CD53 which exhibited the highest score,12.714 (Fig. 4B). These findings indicated that these genes exhibited higher hub degrees and could play critical roles in the development of IgAN.
GO analysis of the module showed significant enrichment in the CC terms of the DEGs were mostly enriched in integral component of plasma membrane, cell surface and plasma membrane.(Fig.5A,6A).MF terms were mainly enriched in superoxide-generating NADPH oxidase, receptor activity and protein binding (Fig.5B,6B). BP terms: innate immune response, integrin-mediated signaling pathway and inflammatory response. (Fig.5C,6C),
The KEGG pathway enrichment analysis revealed that the DEGs were mainly enriched in Tuberculosis, natural killer cell mediated cytotoxicity, Osteoclast differentiation and Staphylococcus aureus infection (Fig.5D,6D).
4. Identification and analysis of hub genes
We exported the STRING data to Cytoscape to construct and visualize the PPI network by implementing the cytoHubba. Thereafter, we implemented the MCC method to evaluate the significance of the genes in the network. The top ten genes, including IL10RA, ITGB2, HCK, C3AR1, CYBB, LAPTM5, FCER1G, CD53, C1QA and TYROBP. Hierarchical clustering of the hub genes was performed as indicated in the clustering heat map (Fig. 7A), these Hub gene could well distinguish the IgAN and control group completely. The biological process analysis of hub genes was performed using BiNGO plug-in showed in Fig.7B.
For further analysis, The microarray data GSE58539 was downloaded from GEO, This dataset contained 17 monocytes samples, including 15 monocytes samples isolated from IgAN patients and 2 monocytes samples isolated from health control group.We used these selected hub genes for analysis, The scatter plot showed that each hub gene was significantly different between IgAN and control group (Fig. 8A). Hierarchical clustering of the hub genes was performed. As indicated in the clustering heat map (Fig. 8B), these Hub gene also could well distinguish the IgAN and control group in monocytes sample.