Table 2.
Model | dN/dSa | Estimates of parametersb | l | Positively Selected Sitesc |
---|---|---|---|---|
M0 (one-ratio) | 0.053 | ω = 0.053 | -22547.97 | not allowed |
M1a (nearly neutral) | 0.084 | p0 = 0.964, (p1 = 0.036) (ω0 = 0.050), (ω1 = 1) |
-22443.90 | not allowed |
M2a (positive selection) | 0.084 | p0 = 0.964, p1 = 0.002, (p2 = 0.033) (ω0 = 0.050), (ω1 = 1), ω2 = 1 |
-22443.90 | not found |
M3 (discrete) | 0.058 | p0= 0.529, p1= 0.355, (p2 = 0.115) ω0 = 0.011, ω1 = 0.073, ω2 = 0.230 |
-22016.13 | not found |
M7 (beta) | 0.063 | p = 0.598, q = 8.411 | -22007.24 | not allowed |
M8 (beta&ω) | 0.220 | p0= 1, (p1 = 0) p = 0.685, q = 1.988, ω = 4.997 |
-19662.03 | 8, 25, 40, 41, 42, 44, 45*, 48, 49, 52*, 53, 55, 56, 57, 60*, 61*, 67, 71, 95, 108, 113, 117, 120, 154, 159, 165*, 170, 173, 175, 193, 197, 207*, 209*, 216, 219, 261, 263*, 265*, 331*, 340, 341, 342, 344, 354* |
a The dN/dS ratio is an average over all sites of bestrophin gene alignments.
b Parameters in parentheses are not free parameters.
c Numbering of amino acid residues corresponds to human bestrophin 1. Positive selection sites with posterior probabilities > 95% are shown in bold.
* The amino acid residues depicted with an asterisk were also found to be implicated in the functional divergence between bestrophin paralogues (see Table 4).