Table 2.
Evaluation of five different circRNA detection methods based on simulated datasets in A. thaliana and O. sativa
| Species | Software | Sensitivity | Precision | F1 score |
|---|---|---|---|---|
| A. thaliana | MapSplice | 0.30 | 0.91 | 0.45 |
| ACFS2 | 0.01 | 0.33 | 0.01 | |
| PcircRNA_finder | 0.85 | 0.90 | 0.87 | |
| CircPlant | 0.92 | 0.99 | 0.95 | |
| CIRI2 | 0.88 | 0.95 | 0.91 | |
| O. sativa | MapSplice | 0.26 | 0.76 | 0.39 |
| ACFS2 | 0.00 | / | 0.00 | |
| PcircRNA_finder | 0.69 | 0.92 | 0.79 | |
| CircPlant | 0.96 | 0.99 | 0.98 | |
| CIRI2 | 0.96 | 0.99 | 0.97 |
Note: Sensitivity = TP/(TP + FN); Precision = TP/(TP + FP); F1 score = 2 × TP/(2 × TP + FP + FN), in which TP (true positive), FP (false positive), and FN (false negative) represent the number of correctly identified circRNAs, the number of incorrectly identified circRNAs, and the number of missing circRNAs, respectively. Sensitivity, precision, and F1 score all range from 0 to 1; higher value means higher accuracy and balance.