Fig. 4
From: Multi-omics integration for neuroblastoma clinical endpoint prediction
Integrated feature set sizes and mixedness. a. Feature set sizes by integration approach for all models. The feature sets identified by INF are more compact than those identified by juxt and rSNF (p-values = 2.453e-08 and 3.803e-09 respectively, Wilcoxon rank sum test). b. and c. Mixedness for the 145 data subset. d. and e. Mixedness for the 498 data subset