Novel immune cross-talk between inflammatory bowel disease and IgA nephropathy

Figures & data

Figure 1. Workflow of this study. All datasets were extracted from Gene Expression Omnibus (GEO) database. The LASSO regression, RFE algorithm, and Boruta algorithm were used to select the optional cross-talk genes. RF model was considered as the best model to predict IgAN using the 10 cross-talk genes. In addition, protein–protein interaction (PPI) and functional enrichment analysis of the DEGs were performed. Graphic created with BioRender.com.

Figure 2. The underlying immune characteristics diversity in IgAN and IBD samples. Differences in pathway activities scored per sample by GSVA in IgAN vs. healthy controls and IBD samples vs. healthy controls. (A) The differences of the Reactome pathway enrichment score between IBD and healthy controls. (B) The differences of the Reactome pathway enrichment score between IgAN and healthy controls.

Table 1. The prediction of RF model in IgAN (GSE93798).

IgAN	RF predict IBD	RF predict non-IBD
Inflammatory bowel disease	30	42
Healthy controls	108	566

Table 2. The prediction of RF model in IBD (tissue samples GSE193677).

IBD	RF predict IBD	RF predict non-IBD
IgA nephropathy	13	1
Healthy controls	9	19

Table 3. The prediction of RF model in IBD (blood samples GSE186507).

IBD	RF predict IBD	RF predict non-IBD
Inflammatory bowel disease	1	42
Healthy controls	61	566

Data availability statement

The code and data can be found on GitHub at the following link: https://github.com/qqyxasdq/Novel-Immune-Cross-talk-Between-IBD-and-IgAN.git