Fig. 3
From: iDESC: identifying differential expression in single-cell RNA sequencing data with multiple subjects
Power comparison of methods considering subject effect in simulated datasets. Evaluation criteria including a sensitivity and b specificity under the p-value threshold of 0.05, and c area under an ROC curve (AUC) to measure the accuracy of identified DE genes under three levels of capture efficiency (δ) and number of cells per subject (\(m\)). d ROC curves and the corresponding AUC scores when \(\beta =0.4, \delta =1, m=50\)