Skip to main content
Fig. 3 | Journal of Translational Medicine

Fig. 3

From: Incorporating knowledge of disease-defining hub genes and regulatory network into a machine learning-based model for predicting treatment response in lupus nephritis after the first renal flare

Fig. 3

Identification of hub genes for LN. A Gene regulatory network constructed for up-regulated DEGs using Cytoscape. B Gene regulatory network constructed for down-regulated DEGs using Cytoscape. C Dot plot reveals the top 10 regulatory hub genes selected by five CytoHubba methods (selected from up-regulated and down-regulated hub genes). The numbers of times selected by the five algorithms are shown (left and right panels). Five methods: MNC, MCC, EPC, DMNC, and Degree. D Heatmap of Wilcoxon’s rank sum test P values derived from the 45 hub genes between LN and control in peripheral blood samples. Color bar indicates –log10-transformed P values that times the sign of regulation. For the blood samples, if the expression is higher in LN, the transformed P value times one; if it is lower in LN, the transformed P value times negative one. The heatmap is separated into up-regulated hub genes (right panel) and down-regulated hub genes (left panel). Insignificant results are colored in light gray. Missing genes are colored in dark gray

Back to article page

Journal of Translational Medicine

ISSN: 1479-5876

Contact us