This approach provided us with a range of lineage values to examine the sensitivity of our statistical inferences to taxonomy.
We implemented a phylogenetic logistic regression between the VS and UVS condition in each avian species and the respective ω-lineage value.
Phylogenetic analyses of F gene nucleotide sequences of contemporary HRSV and HMPV strains (generated from samples collected since 2007) were constructed as described above for identification of bootstrap-supported monophyletic lineages (values ≥70%) using MEGA v4.0.
The boxplots (Fig. 3a) depict that sw1 ω-lineage values are strongly correlated with the VS/UVS condition in birds, with the accelerated lineages being the most-strongly linked with UVS sensitivity: VS log-odd score = 4.44 – 79.91ω (p-value = 0.046; Wald z-statistics, Additional file 6).
Additionally, a robust principal component analysis of ω-lineage values was conducted following the Hubert (2005) approach [ 82], in order to assess the avian opsin outlier map, which summarizes those lineages that show unexpected ω-tendencies in the evolutionary dimensional space of the studied opsins.
Applying a mutation rate of 0.01 s/s/Myr per lineage, this value corresponds to a time since the expansion of 54,744 years, with confidence intervals between the present and 69,890 years ago.
Doing so, we observed a significant difference in the rate of evolution among the different categories in the hominid lineage (p-value = 0.017) but no difference was observed in the rodent lineage (p-value = 0.59, Fig. 3B).
To test this, for each lineage, P values were computed with a 2×2 contingency table using the Fisher exact test.
Furthermore, the PPAR genes were spitted into three lineages (support value = 100%).
Therefore, significance of the gsi statistic indicates exclusive ancestry of lineages, whereas the value of the gsi measures the degree of lineage divergence.
Lineage-specific values of each property were calculated for all TFs in expressed in our dataset.
