Table 3 Prediction performance on the training, testing, and independent datasets
From: Genome-scale MicroRNA target prediction through clustering with Dirichlet process mixture model
Dataset | Performance | |||
|---|---|---|---|---|
Sensitivity | Specificity | Accuracy | MCC | |
Training 75% of CLASH, iPAR, and negative | 0.78 | 0.86 | 0.82 | 0.64 |
Testing 25% of CLASH, iPAR, and negative | 0.77 | 0.86 | 0.82 | 0.64 |
Validation-1 CLEAR-CLIP | 0.80 | – | 0.80 | – |
Validation-2 mirTarbase (validated) | 0.61 | – | 0.61 | – |
Validation-3 mirTarbase (predicted) | 0.62 | – | 0.62 | – |