Table 5.
Accuracy versus Exon Type
| Programs | Exon type | |||
| initial (152) | internal (601) | terminal (152) | single (43) | |
| FGENES | 0.64 | 0.79 | 0.66 | 0.58 |
| (0.55) | (0.78) | (0.58) | (0.83) | |
| GeneMark.hmm | 0.40 | 0.78 | 0.52 | 0.30 |
| (0.48) | (0.72) | (0.51) | (0.65) | |
| Genie | 0.49 | 0.76 | 0.61 | 0.70 |
| (0.45) | (0.82) | (0.57) | (0.68) | |
| Genscan | 0.57 | 0.87 | 0.67 | 0.63 |
| (0.71) | (0.76) | (0.73) | (0.83) | |
| HMMgene | 0.68 | 0.78 | 0.70 | 0.77 |
| (0.72) | (0.83) | (0.73) | (0.79) | |
| Morgan | 0.35 | 0.55 | 0.36 | — |
| (0.35) | (0.46) | (0.36) | ||
| MZEF | — | — | — | — |
-
The HMR195 dataset was partitioned according to the type of the annotated exons. The number in parenthesis in the header of each column represents the number of actual exons belonging to each partition. For each program, CRa (the upper number) and CRp (the number in parentheses), are averaged over all sequences belonging to that particular partition.
