Fig. 5From: Capturing the differences between humoral immunity in the normal and tumor environments from repertoire-seq of B-cell receptors using supervised machine learningReceiver Operating Characteristic (ROC) curves showing tissue-level classification performance. Each figure illustrates the comparison of ROC-curves of (a) models using average score from sequence-independent features, (b) those using median score, (c) those using mode score, (d) ensemble model and clonal entropy, and (e) ensemble model and Ostmeyer’s modelBack to article page