Table 2. Tests for positive selection among Sdr4L codons of using site models.

Model	InL	2ΔL	Estimates of parametersa	Positively selected sitesb
M0 (one ratio)	-8461.241		ω = 0.16456	None
M3 (discrete)	-8244.734	433.014 (M0 vs M3)	p0 = 0.55168, p1 = 0.33358, p2 = 0.11474, ω0: 0.05766, ω1:0.23719, ω2: 0.66151	None
M7 (beta)	-8240.96		p = 0.73100, q = 2.96832	Not allowed
M8 (beta & ω)	-9418.204	1177.244 (M7 vs M8)	p0 = 0.99999, p = 0.36186, q = 1.15287, p1 = 0.00001, ω = 2.06669	5Q, 74 P**, 77 P*, 78 P, 158V**, 174 E*, 186 H, 193 D**, 194 V*, 195 G, 203 A*, 204A**, 205A**, 206 P*, 249 P, 302 S**, 303 S*

Note

^a Number of parameters in the ω distribution (ratios of non-synonymous (dN or Ka) versus synonymous (dS or Ks) mutations).

^b Positively selected sites are inferred from posterior probabilities > 95% (*) or 99% (**).

Sites implicated in Type I and Type II divergence are shown in bold. Codon (amino acid) positions are based on rice OsSDR4 protein.