Table 1 Overall performance of REGAL on A. thaliana. We tested the performance of REGAL on known edited and unedited sites from three mitochondrial genomes. The results from A. thaliana were obtained after 100 iterations of cross-validation using 100 edited and 100 unedited sites per testing data set. We report the range of values as obtained from the cross-validation for this genome.
From: Genetic algorithm learning as a robust approach to RNA editing site prediction
Known Edited Sites Total: 100 | Known Unedited Sites Total: 100 | ||
|---|---|---|---|
Predicted Edited Site | True positive 81 (± 2.4) | False positive 9 (± 0.4) | Sensitivity: 0.81 (0.80–0.82) Specificity: 0.91 (0.90–0.92) |
Predicted Unedited Site | False negative 19 (± 0.5) | True negative 91 (± 5.4) | Accuracy: 0.86 (0.86–0.90) |