Table 4.
Parameter estimates and likelihood ratio tests for the site models
| Model | p a | LnLb | Estimates of Parameters | 2△lc | df | p | Positively selected sites |
|---|---|---|---|---|---|---|---|
| M0 (one ratio model) | 1 | -47935.395 | ω=0.143 | - | - | - | None |
| Site-specific models | |||||||
| M1: Neutral (k=2) | 1 | -47271.650 | p 0=0.774, (p 1=0.226) | - | - | - | Not allowed |
| M2 : Selction (k=3) | 3 | -47271.650 | p 0=0.774, p 1=0.048, (p 2=0.178), ω2=1.000 | M2 vs M1: 0.000 | 1 | 1.000 | None |
| M3: discrete (K=5) | 5 | -46728.169 | p 0=0.244, p 1=0.389, p 2=0.263, p 3=0.094, p 4=0.011, ω0=0.010, ω1=0.085, ω2=0.317, ω3=0.799, ω4=6.007 | M3 vs M0: 2414.452 | 4 | 0.000 | 1 (p>0.95) |
| M7: beta | 2 | -46761.572 | p=0.513, q=1.910 | - | - | - | Not allowed |
| M8: beta & w>1 | 4 | -46732.126 | p 0=0.947, p=0.589, q=2.892, (p 1=0.053) ω=1.293 | M8 vs M7: 58.891 | 2 | 0.000 | 6 (p>0.95), 3 (p>0.95) |
aThe number of free parameters for the ω ratios
bLikelihood of the model
c2(l1-l0)