Fig. 6
From: Impact of short-read sequencing on the misassembly of a plant genome
Important features in Model 37 and Model 38. a Model 37 is for classifying HC_M2/M3 and HC_C2/C3 to assess the features important for predicting misassembled HC regions. b Mode 38 is for classifying HC_M2/M3 and BG_M2/M3 to assess if misassembled HC and BG regions have distinguishable features. Bar height: importance value in the Random Forest model. Blue, red, green: median values of features are higher in HC_M2/M3, HC_C2/C3, and BG_M2/M3, respectively. The distributions of feature values were shown in Fig. S5 and Fig. S6 for Model 37 and Model 38, respectively