|  | A | O | C | Asia1 | SAT1 | SAT2 | SAT3 |  | A | O | C | Asia1 | SAT1 | SAT2 | SAT3 |
---|
 | 129 | 47 | 49 | 8 | 9 | 10 | 3 | 3 | 83 | 6 | 31 | 16 | 26 | 1 | 2 | 1 |
A | 47 |
47
| Â | Â | Â | Â | Â | Â | 6 |
6
| Â | Â | Â | Â | Â | Â |
O | 48 | Â |
48
| Â | Â | Â | Â | Â | 31 | Â |
31
| Â | Â | Â | Â | Â |
C | 8 | Â | Â |
8
| Â | Â | Â | Â | 16 | Â | Â |
16
| Â | Â | Â | Â |
Asia1 | 9 | Â | Â | Â |
9
| Â | Â | Â | 26 | Â | Â | Â |
26
| Â | Â | Â |
SAT1 | 9 | Â | Â | Â | Â |
9
| Â | Â | 1 | Â | Â | Â | Â |
1
| Â | Â |
SAT2 | 4 | Â | 1 | Â | Â | Â |
3
| Â | 2 | Â | Â | Â | Â | Â |
2
| Â |
SAT3 | 4 | Â | Â | Â | Â | 1 | Â |
3
| 1 | Â | Â | Â | Â | Â | Â |
1
|
- The composition of the different serotype FMDV strains in our two datasets (columns 2 and 10). The LOOCV genotype prediction results on the first dataset are in row 2 from columns 3 to 9 and more details in rows 3–9 and columns 3–9 as a confusion matrix; the independent testing results on the second dataset are in row 2 from columns 11–17 and more details in rows 3–9 and columns 11–17 as a confusion matrix. Numbers in bold denote the correct genotype predictions.