Research Article
Predicting Potential PRRSV-2 Variant Emergence through Phylogenetic Inference
Table 1
The best fit model for each success aspect and predicted period.
| Success aspect | Measure of success | Scenario (model) | Model p-value (LRT) | Predictor | p-value | Coefficient | Raw OR | Adjusted OR | Predictor’s lower upper quartiles | Training set | Test set | Concordance | Sensitivity | Specificity | PPV† | NPV†† | F1 score | Balanced accuracy |
| Population expansion | Absolute increase in number of taxa | 12-t-12 | 0.002 | Branch length | 0.098 | −1,423.0 | 7.9 × 10−63 | 0.12 | 0.00–0.02 | 0.875 | 0.75 | 0.58 | 0.38 | 0.88 | 0.500 | 0.667 | Average clade rate | 0.808 | −200.0 | 1.3 × 10−87 | 0.4 | 0.00–0.01 | LBI | | 8.1 | 3.3 × 103 | 12.12 | 0.03–0.333 | Amino acid distance | 0.159 | 20.9 | 1.1 × 109 | 3.15 | 0.020–0.075 | N-gly similarity | 0.410 | 1.4 | 4.0 | 1.79 | 0.333–0.750 | 24-t-24 | 0.001 | Branch length | 0.120 | −220.4 | 1.8 × 10−96 | 0.08 | 0.002–0.013 | 0.810 | 1.00 | 0.50 | 0.40 | 1.00 | 0.571 | 0.750 | Ancestral rate | 0.446 | 268.9 | 6.1 × 10116 | 3.65 | 0.002–0.007 | LBI | | 8.9 | 7.2 × 103 | 12.67 | 0.000–0.286 | Amino acid distance | 0.652 | 7.7 | 2.2 × 103 | 1.42 | 0.030–0.075 | N-gly similarity | 0.939 | 0.2 | 1.2 | 1.09 | 0.333–0.750 |
| Spatial distribution | Relative increase in maximum between-county geodetic distance | 12-t-12 | 0.006 | Branch length | 0.587 | 19.9 | 4.3 × 108 | 1.34 | 0.003–0.018 | 0.729 | 0.75 | 0.42 | 0.30 | 0.83 | 0.429 | 0.583 | Ancestral rate | | −898.7 | 0 | 0 | 0.001–0.007 | LBI | 0.116 | 5.8 | 3.5 × 102 | 6.04 | 0.026–0.333 | Amino acid distance | 0.725 | 5.9 | 3.5 × 102 | 1.38 | 0.020–0.075 | N-gly similarity | 0.081 | 2.8 | 1.7 × 101 | 3.24 | 0.333–0.750 | 12-t-24 | 0.029 | Branch length | | −170.7 | 7.5 × 10−75 | 0.1 | 0.003–0.017 | 0.792 | 0.50 | 0.83 | 0.50 | 0.83 | 0.500 | 0.667 | Ancestral rate | 0.136 | −309.4 | 4.3 × 10−135 | 0.17 | 0.002–0.008 | LBI | 0.293 | 5.5 | 2.5 × 102 | 5.13 | 0.026–0.322 | Amino acid distance | | 53.4 | 1.6 × 1023 | 14.47 | 0.025–0.075 | N-gly similarity | 0.620 | 0.8 | 2.2 | 1.39 | 0.333–0.750 |
| Genetic diversity | Absolute increase in 95th percentile pairwise nucleotide distance | 12-t-12 | <0.001 | Branch length | | 108.1 | 8.6 × 1046 | 4.92 | 0.003–0.018 | 0.917 | 1.00 | 0.58 | 0.44 | 1.00 | 0.615 | 0.792 | Ancestral rate | 0.603 | −141.6 | 3.2 × 10−62 | 0.42 | 0.001–0.007 | LBI | 0.758 | −1.2 | 2.9 × 10−1 | 0.69 | 0.026–0.333 | Amino acid distance | 0.909 | 2.1 | 7.9 | 1.12 | 0.020–0.075 | N-gly similarity | 0.779 | 0.6 | 1.7 | 1.27 | 0.333–0.750 | 24-t-24 | 0.006 | Branch length | 0.119 | 62.1 | 9.4 × 1026 | 2.07 | 0.002–0.013 | 0.857 | 1.00 | 0.33 | 0.33 | 1.00 | 0.500 | 0.667 | Ancestral rate | 0.880 | −46.6 | 5.9 × 10−21 | 0.8 | 0.002–0.007 | LBI | 0.958 | −0.2 | 8.2 × 10−1 | 0.95 | 0.000–0.286 | Amino acid distance | | 42.5 | 2.9 × 1018 | 6.78 | 0.030–0.075 | N-gly similarity | 0.535 | 1.8 | 5.8 | 2.09 | 0.333–0.750 |
|
|
, †positive predictive value, ††negative predictive value. |