Skip to main content

Table 1 Parameter estimates and LRTs in the mutation-selection models

From: Selection on synonymous codons in mammalian rhodopsins: a possible role in optimizing translational processes

Model

np

lnl

p-value of LRT

к

ω

Ï€C*

Ï€G*

Ï€T*

Ï€A*

M0-set (alternative FMutSel-M0 tested against null FMutSel0-M0)

FMutSel0-M0

40

−6015.3

N/A

3.22

0.075

0.45

0.29

0.13

0.13

FMutSel-M0

81

−5878.3

3.14 × 10 −36 (df = 41)

2.94

0.050

0.19

0.20

0.20

0.42

M3-set (alternative FMutSel-M3 tested against null FMutSel0-M3

FMutSel0-M3

42

−5860.4

N/A

3.28

ω0 = 0.012 ω1 = 0.431 p0 = 83.0%, p1 = 17.0%

0.45

0.29

0.13

0.13

FMutSel-M3

83

−5722.8

1.99 × 10 −36 (df = 41)

3.03

ω0 = 0.006, ω1 = 0.272, p0 = 81.6%, p1 = 18.4%

0.19

0.20

0.18

0.43

  1. np is the number of parameters in the model, lnl is the log likelihood score, p-value is the result of likelihood ratio tests (LRTs), df is the degrees of freedom in LRTs, к is the transition/transversion ratio, ω is the non-synonymous/synonymous substitution ratio, πN* (N = C, G, T, A) is the parameter of mutational bias for C, G, T, A, respectively.