Fig. 6
From: Deep learning improves the ability of sgRNA off-target propensity prediction
Leave-one-sgRNA-out comparison of sgRNA off-target efficacy prediction with auROC, auPRC, Pearson correlation coefficient, and Spearman correlation coefficient. The 29 test results were plotted as the violin plots. The two figures above show the performance comparison results in classification schema and the two figures below shows the performance comparison results in regression schema. It is worth noting that the article of DeepCrispr did not offer the detailed test results of leave-one-sgRNA-out validation, as a result we did not draw the corresponding violin plot. It is anticipated that this comparison process can further improve upon future works