Schematic of the nested CV procedure and performance comparison. a The nested CV procedure used for CRISPRi, CRISPRa, Cas9, and Cas12a. TR: Training; TS: Test. b–e The plots depict AUC values for each algorithm across CRISPR-Cas systems. AUC values for subsets as well as complete training data are shown. The x-axis of each plot denotes subset and complete training data, and the y-axis denotes AUC values. f–i ROC curves depict the algorithms’ performance on the complete training data. In the ROC plots, the x-axis denotes the average False Positive rate (FP) and the y-axis denotes the average True Positive rate (TP).