Identification of Nephrogenic Therapeutic Biomarkers of Wilms Tumor Using Machine Learning

Figure 2

Identification of differential genes in Wilms tumor. (a) Venn diagram, (b) error rate, and variation importance of the 48 differentially expressed genes using dimensionality reduction analysis of random survival forest. (c) Target genes with a relative importance > 0.3 of the random survival forest. Heat map of the eight differentially expressed genes finally identified in normal and tumor tissues in (d) TARGET, (e) GSE66405, and (f) GSE73209.

(a)

(b)

(c)

(d)

(e)

(f)

(g)