Table 2.
Accuracy of GeneID in theAdh Region
| Base level | Exon level | CPU time (sec) | Memory (MB) | ||||||
| Sn | Sp | Sn | Sp | ME | WE | ||||
| (std1) | (std3) | (std1) | (std3) | (std1) | (std3) | ||||
| GeneID,submitted (1) | 0.48 | 0.84 | 0.27 | 0.29 | 54.4 | 47.9 | 74 | ∼500 | |
| GeneID, submitted (2) | 0.86 | 0.82 | 0.59 | 0.34 | 21.0 | 48.0 | 74 | ∼500 | |
| GeneID,current | 0.96 | 0.92 | 0.70 | 0.62 | 11.0 | 17.0 | 83 | 18.11 | |
-
The std1 annotation data set was used to evaluate sensitivity; the std3 annotation data set to evaluate specificity, as in GASP1 (seeReese et al. 2000). Discrepancies between the accuracy of the submitted predictions, both the initial ones (1) and the corrected (2), and the accuracy of the predictions obtained with the current version ofGeneID are due to a number of errors during the process of generating the submitted predictions (see Discussion). The decrease in the amount of memory required to obtain the predictions is due to algorithmic developments occurring after GASP1.
