Figure 2From: Selecting a single model or combining multiple models for microarray-based classifier development? – A comparative analysis based on large and diverse datasets generated from the MAQC-II projectThe ensemble models vs. the NCTR nominated models. A pair-wise t-test was applied to the MCCs obtained from the ensemble models and the NCTR nominated models. (P-value = 0.039 if two random endpoints, i.e., I and M, were excluded).Back to article page