Table 1.

Tests for positive selection among codons of SmSABATH genes using site models.

Model	np	lnL	Estimates of Parameter 1		2ΔlnL	Positive Selection Sites 2
			Frequency	dN/dS
M0 (one-ratio)	60	−24,100.940606		0.32227		None
M3 (discrete)	64	−23,631.868500	p0 = 0.15469 p1 = 0.41191 p2 = 0.43340	ω0 = 0.06316 ω1 = 0.25310 ω2 = 0.65803	938.1442 (M3 vs. M0) **	Not allowed
M1a (nearly neutral)	61	−23,766.855086	p0 = 0.53255 p1 = 0.46745	ω0 = 0.23043 ω1 = 1.00000		None
M2a (positive selection)	63	−23,766.855086	p0 = 0.53255 p1 = 0.30714 p2 = 0.16031	ω0 = 0.23043 ω1 = 1.00000 ω2 = 1.00000	0 (M2a vs. M1a)	Not allowed
M7 (beta)	61	−23,622.773935	p = 1.10441 q = 1.67064			None
M8 (beta & ω)	63	−23,622.775486	p0 = 0.99999 p = 1.10448 q = 1.67080 p1 = 0.00001	ω = 5.70606	0.00308 (M8 vs. M7)	181 Q 432 V 434 G 460 S 476 H 479 C

Note: * p < 0.05 and ** p < 0.01 (x² test). ¹ ω was estimated under models; p and q were the parameters of the beta distribution. ² The number of amino acid sites estimated to have undergone positive selection. np was the number of parameter. lnL was the values of log-likelihood. dN/dS was non-synonymous/synonymous. 2ΔlnL was the value of twice the log likelihood difference between models.