Table 2 |
||
|
Identification of splice junctions |
||
| Known trans-splicing regions |
Known coding regions |
|
|
|
||
| Predicted splice sites |
True positive |
False positive |
| 17 |
2.5 |
|
| (10-22) |
(0-4) |
|
| Predicted nonsplice sites |
False negative |
True negative |
| 4.5 |
13.9 |
|
| (1-8) |
(11-16) |
|
| Sensitivity: 0.80 |
Specificity: 0.85 |
|
| (0.71-0.93) |
(0.75-1.00) |
|
| Accuracy: 0.82 |
||
| (0.74-0.93) |
||
|
|
||
|
The overall performance of the method in identifying splice junctions was determined by comparing the number of known splice junctions that were identified by the method in known trans-splicing regions versus those in known coding regions. These results are from tenfold cross-validation, and each test dataset had on average 21.5 known trans-splicing regions and 16.4 known coding regions. |
||
|
Gopal et al. Genome Biology 2005 6:R95 doi:10.1186/gb-2005-6-11-r95 |
||
