Dataset | Sub-family | Total | Predicted | | | | Accuracy | | | | |
---|
This paper | [9] | [23] | [15] | [16] | [18] | [22] | [25] |
---|
D167
| Acetylcholine | 31 | 29 |
93.54
|
93.54
| 100 | 67.74 | 90.3 | 93.6 | 93.3 | 96.7 |
Adrenoceptor | 44 | 44 |
100
|
81.81
| 100 | 88.64 | 86.4 | 100 | 100 | 100 |
Dopamine | 38 | 37 |
97.36
|
97.36
| 94.74 | 81.58 | 78.9 | 92.1 | 94.7 | 92.1 |
| Serotonin | 54 | 54 |
100
|
81.48
| 98.15 | 88.89 | 79.6 | 98.2 | 100 | 100 |
- Accuracy is calculated as (TP + TN)/(TP + FN + TN + FP) here and throughout.
- “Total” is the number of sequences observed in a sub-family, “Predicted” is the number of sequences correctly predicted by our method. The accuracy values for the other methods are directly taken from their articles. The data in bold depicts the results using HMM-ModE with HMMER3 and HMMER2.