Computational protocol: Molecular Evolution of the Infrared Sensory Gene TRPA1 in Snakes and Implications for Functional Studies

Similar protocols

Protocol publication

[…] Deduced amino acid sequences of TRPA1 and TRPV1 were aligned with CLUSTALW and modified with BioEdit . The nucleotide alignments were generated according to amino acid alignments. The best-fitting models for the TRPA1 and TRPV1 DNA alignments were separately selected by AIC and implemented in MrModelTest2.3 . The GTR+I+Γ model was chosen as the best-fitting model for both the TRPA1 and the TRPV1 genes. ML analyses were implemented using RAxML 7.0.3 with 1000 rapid bootstrap replicates. Bayesian analyses were performed in MrBayes 3.1.2 . Two MCMC runs were performed with one cold and three heated chains (temperature set to 0.2) for 10 million generations and sampled every 500 generations. The first 25% of sampled trees were discarded as the burn-in. Similar topologies and posterior clade probabilities from the two runs were observed. We also analyzed the protein alignments of TRPA1 and TRPV1 following the same methodology using the JTT+Γ as the amino acid substitution model.Because TRPA1 is the receptor molecule in the pit organ, we suspect it may experience natural selection within pit-bearing snakes. To visualize selective variation among the different amino acid sites of the TRPA1 gene, we performed a sliding window analysis using the program SWAAP 1.0.2 , comparing pit-bearing snakes with other non-pit snakes and vertebrates. The window size and the step size were set at 150 bp and 15 bp, respectively. Values of ω were estimated following Nei and Gojobori . To test for evidence of positive selection in TRPA1 across certain groups, we implemented site models with the CODEML program in PAML 4.4 , comparing Model M1a with Model M2a using the LRT test. We also applied a two ratio model test with CODEML to the entire TRPA1 dataset, with one omega parameter assigned to all pit-bearing snake clades and the other assigned to the remaining lineages (non-pit snakes and other vertebrates), to detect if the TRPA1gene of all pit-bearing snake species experienced divergent patterns of selection compared to non-pit species and other vertebrates.After finding evidence of significant variation in ω of TRPA1 across different groups, we applied the GABranch method to investigate how this variation was distributed across the branches of the TRPA1 tree. This analysis was conducted using downloadable HyPhy script ( implemented in HyPhy version 1.0 . The nucleotide model was specified as GTR; otherwise, the default GABranch configuration was used. Unlike the free-ratio model implemented in PAML, the GABranch method does not calculate the ω-value precisely for a given branch but assigns it to a ω-category, avoiding the overparameterization problem of the free-ratio model (PAML manual) . Although the GABranch method is useful to assign branches into similar selective categories, the estimated ω-value is not ideal. Therefore, for those branches assigned to ω-categories exceeding (or nearly exceeding) one by the GABranch method, we used the branch models in PAML 4.4 to recalculate the ω-value and tested whether the ω-value of a given branch was significantly higher than one (evidence for positive selection). In addition, as a “negative control”, we also applied the GABranch analysis to the TRPV1 gene. Because TRPV1 belongs to the TRP channel family, like TRPA1, but is not involved in infrared detection , we had expected to observe a different pattern of selective pressure variation across branches of the TRPV1 tree.In order to identify amino acid changes that may be responsible for infrared sensitivity in pit-bearing snake TRPA1 proteins, we adopted the function-based method used by Yokoyama et al. . This method aims to identify amino acid residues that are totally conserved in the “control group” (functionally important) but diverged to different states in the “target group” (functional divergence). In this case, the control group is composed of the TRPA1 proteins of all sampled non-pit snakes, which we assume are not infrared-sensitive; the target group is the TRPA1 proteins of all sampled pit-bearing snakes. Because the infrared sensitivity of pit vipers is 5–10-fold higher than pythons or boas , we also set the pit vipers as the target group, while all other snakes were set as the control group, to identify amino acid changes that may contribute to further enhancing infrared sensitivity. […]

Pipeline specifications

Software tools Clustal W, BioEdit, MrModelTest, RAxML, MrBayes, PAML, HyPhy
Application Phylogenetics
Chemicals Calcium