Figure 3
From: Efficient discovery of responses of proteins to compounds using active learning
Evaluation of prediction methods for held out targets or compounds with increasing amounts of randomly selected training data. Ten random sequences of experiments were used to select data from compounds or targets not held out and used for training regression models. After experiments were chosen, a ROC curve was constructed by gradually raising the threshold on the predicted assay score at which an experiment was considered to be positive. The mean and standard error of the area under the ROC curve for prediction of positive experiments from held out targets or compounds are plotted for held out compounds (blue) or targets (cyan).