Fig. 3
From: Comparative epigenomics by machine learning approach for neuroblastoma
Prediction power of probe groups. (A) Effect of probe groups in prediction of group A. Blue line represents the random-sampling of probes. Orange line shows select percentile method of scikit-learn with default setting. (B) Group B version of Fig. 3A). (C) Relationship between the variance of β and obs/sim f1-score for group A. (D) Relationship between the variance of β and obs/sim f1-score for group B