Table 3.
Comparison of selective pressure by transmembrane domain.
| TM1 | TM2 | TM3 | TM4 | TM5 | TM6 | TM7 | non-TM | CDS | |
| Length (amino acids) | 26 | 21 | 19 | 24 | 25 | 19 | 24 | 110 | 268 |
| Mean pairwise dN/dSa | 0.330 | 0.276 | 0.457 | 0.574 | 0.437 | 0.513 | 0.282 | 0.304 | 0.325 |
| TM vs loops p-value (2 -tailed t test)b | 9.9E-04 | 0.81 | 2.1E-12 | 7.6E-21 | 1.2E-18 | 2.5E-10 | 1.00 | ||
| TM vs CDS p-value (2 -tailed t test)b | 0.45 | 0.025 | 2.1E-08 | 5.9E-16 | 7.9E-12 | 6.0E-07 | 0.0024 | ||
| Number of positively selected sites (p < 0.1)c | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 2 | 4 |
| Number of negatively selected sites (p < 0.1)c | 21 | 16 | 12 | 15 | 15 | 14 | 19 | 80 | 192 |
a The mean ratios of non-synonymous substitutions per non-synonymous site (dN) to synonymous substitutions per synonymous site (dS) were calculated for each transmembrane region, the aggregate of the non-transmembrane regions, as well as the entire OR coding sequence. dN/dS ratios were calculated only for pairwise comparisons in which mutational saturation had not been reached. A constant set of 262 informative pairs was used for all calculations.
b Probabilities that the average dN/dS ratio for the particular TM domain is no different than the average for the entire CDS or the non-TM domains, using a two-tailed t test.
c Number of sites with probabilities greater than 0.9 (p < 0.1) of being positively or negatively selected for each domain. Ancestral sequences were reconstructed, and at every non-constant site, dN and dS were calculated. If dN < (or >) dS, a p-value derived from a two-tailed binomial distribution was used to assess significance.