Table 2.
Evaluation of the goodness of fit for different models of codon evolution and estimated parameter values
| Model | LnL | AIC | ∆AIC | Parameters |
|---|---|---|---|---|
| Mimu-DRB | ||||
| M0 — one ω | −1,662.8 | 3,327.7 | 306.9 | ω = 0.82 |
| M7 — nearly neutral with β | −1,517.7 | 3,055.4 | 34.8 | |
| M8 — positive selection with β (ω 0 ≤ 1, ω 1 > 1) | −1,499.4 | 3,020.8 | Best | p 0 = 0.75, p 1 = 0.25, ω 1 = 2.92 |
| Mimu-DQB | ||||
| M0 — one ω | −1,773.0 | 3,548.0 | 433.1 | ω = 0.65 |
| M7 — nearly neutral with β | −1,587.0 | 3,194.1 | 79.2 | |
| M8 — positive selection with β (ω 0 ≤ 1, ω 1 > 1) | −1,546.5 | 3,114.9 | Best | p 0 = 0.94, p 1 = 0.05, ω 1 = 5.51 |
ω — dN/dS; nearly neutral with beta — for all sites ω ≤ 1 and the beta distribution approximates ω variation; positive selection — a proportion of sites evolves with ω > 1; p 0 — proportion of sites with ω ≤ 1, p 1 — proportion of positively selected sites (ω > 1), ω 1 — estimated value of ω for sites under positive selection
AIC Akaike information criterion, ΔAIC difference between the value of the AIC of a given model and the best model