Skip to main content
. 2018 Apr 16;8:6005. doi: 10.1038/s41598-018-24298-y

Table 1.

Performance comparison of DeepRed with separate samples, GIREMI, RNAEditor and Prediction methods on U87 data.

Method Accuracy Sensitivity Specificity GM Positive predict value F1 score Validation rate Misclassification rate
DeepRed 93.23% 79.66% 99.25% 88.92% 97.92% 88.38% 29.01% 0.62%
Separate 76.04% 83.05% 72.93% 77.83% 57.65% 77.66% 29.34% 21.56%
GIREMI 70.31% 6.78% 98.50% 25.84% 66.67% 12.69% 14.81% 7.41%
RNAEditor 89.58% 69.49% 98.50% 82.73% 95.35% 81.49% 40.59% 1.98%
Prediction 63.54% 15.25% 84.96% 36.00% 31.03% 25.86% 7.83% 17.39%