Table 2.
Model | np | estimates of parameters | lnL | LRT pairs | df | 2flnL | positively selected sites (BEB) |
---|---|---|---|---|---|---|---|
M0:one ratio | 1 | ω0:one | -21586.13 | ||||
M3:discrete | 5 | p0=0.31,p1=0.46,p2=0.23,p0=0.05,p1=0.43,ω2=1.23 | -20801.49 | M0/M3 | 4 | 1569.28*** | |
M1a:neutral | 2 | p0=0.56,p1=0.44,p0=0.17,p1=1.00 | -20956.12 | ||||
M2a:selection | 4 | p0=0.56,p1=0.34,p2=0.10,p0=0.17,p1=1.00,p2=1.00 | -20956.12 | M1a/M2a | 2 | 0 | 3 site p<0.01: 159Q,284E,337E;4 site p<0.05: 238S, 260K, 291E, 305D |
M7:beta | 2 | p=0.48,q=0.63 | -20817.70 | ||||
M8:beta&0 | 4 | p0=0.92,p=0.56,q=0.88,(p1=0.08),ω0.08) | -20780.34 | M7/M8 | 2 | 74.72*** | 8 site p<0.01:159Q, 202E, 238S, 260K,284E, 291E,305D,337E; 3 site p<0.05: 211Y,359A,364T |
Fr:free ratios | 92 | see Figure | -21416.49 | M0/Fr | 91 | 339.28*** |
Selection analysis by three types of models was performed using Codeml implemented in PAML. np: number of free parameters, InL: log likelihood. LRT: likelihood ratio test. df: degrees of freedom. 2∆LnL: twice the log-likelihood difference of the models compared. The significant tests at 5% cutoff are labeled with *, and those at 1% cutoff are labeled with ***.