TABLE 3.
Group | Gene | Characterizationa | NSyn | Syn | No. of codons | Ratio | Significanceb |
---|---|---|---|---|---|---|---|
Combined characterization of exposure status at codons | |||||||
Major | VP2 | B | 62 | 1,172 | 162 | 0.053 | |
Minor | VP2 | B | 22 | 340 | 162 | 0.065 | NS |
Major | VP3 | B | 82 | 973 | 130 | 0.084 | |
Minor | VP3 | B | 19 | 268 | 130 | 0.071 | NS |
Major | VP1 | B | 138 | 1,117 | 157 | 0.123 | |
Minor | VP1 | B | 25 | 318 | 157 | 0.071 | NSB χ2 = 3.65; P value = 0.056 |
Major | VP2 | E | 254 | 777 | 97 | 0.363 | |
Minor | VP2 | E | 123 | 248 | 97 | 0.496 | SB χ2 = 9.64; P value = 0.0019 |
Major | VP3 | E | 257 | 847 | 107 | 0.294 | |
Minor | VP3 | E | 100 | 246 | 107 | 0.406 | SB χ2 = 5.19; P value = 0.023 |
Major | VP1 | E | 353 | 934 | 125 | 0.378 | |
Minor | VP1 | E | 180 | 291 | 125 | 0.619 | SB χ2 = 18.5; P value = 0.000017 |
Joint characterization of exposure status at codons | |||||||
Major | VP2 | R | 42 | 414 | 53 | 0.101 | |
Minor | VP2 | R | 29 | 124 | 53 | 0.234 | SB χ2 = 9.64; P value = 0.0019 |
Major | VP3 | R | 82 | 563 | 69 | 0.146 | |
Minor | VP3 | R | 33 | 153 | 69 | 0.216 | NS |
Major | VP1 | R | 166 | 674 | 91 | 0.246 | |
Minor | VP1 | R | 85 | 220 | 91 | 0.386 | SB χ2 = 8.12; P value = 0.0044 |
Major | VP2 | R+S | 192 | 325 | 40 | 0.591 | |
Minor | VP2 | R+S | 89 | 111 | 40 | 0.802 | NS |
Major | VP3 | R+S | 174 | 303 | 37 | 0.574 | |
Minor | VP3 | R+S | 67 | 92 | 37 | 0.728 | NS |
Major | VP1 | R+S | 187 | 260 | 34 | 0.719 | |
Minor | VP1 | R+S | 95 | 71 | 34 | 1.338 | SB χ2 = 10.9; P value = 0.00094 |
R, exposed by the rolling-only model; R+S, exposed by the rolling+structural model.
NS, not significant; SB, significant after Bonferroni correction; NSB, not significant after Bonferroni correction.
We show the number of nonsynonymous changes (NSyn), the number of synonymous changes (Syn), and the number of codons for each class of codons (exposed [E] or buried [B]) in major- and minor-group lineages for each gene, along with the ratio of nonsynonymous to synonymous changes. The significance is based on a chi-square test of the 2-by-2 contingency table of nonsynonymous and synonymous changes for these two rows. Note that we did not include the five sites identified only by the structural model because they could not have produced a significant interaction.