Fig. 4

scMTD improves the differential expression analysis in the Chu (Time Course) dataset. a Plot shows the receiver operating characteristic (ROC) curves and area under the curve (AUC) for the accuracy of DEG predictions of the raw data and the imputed data. b Box plots show the LFC distributions of the top 100 ground-truth DEGs in the bulk RNA-seq data and (raw and imputed) scRNA-seq data. c Plot shows the enriched GO terms related to the molecular function only detected in the scMTD imputed data, but not in the raw data. d Box plots show the expression distributions of 15 marker genes in the raw data and scMTD imputed data, and the y-axis represents the logarithm transformation of expressions