Similar protocols

Pipeline publication

[…] []. The mean numbers of nonsynonymous substitutions per nonsynonymous site (dN) and that of synonymous substitutions per synonymous site (dS) were calculated for excluding identical sequences of DQB variants. We estimated nucleotide diversity (πS) among all the sequences in each management unit or across the entire population. The divergence time of the Japanese black bear DQB variants was estimated using synonymous substitutions alone. We assumed that the maximum genetic distance at synonymous sites roughly corresponded to the TMRCA of DQBs. These genetic distance parameters (πS, dN, and dS) were calculated using the modified Nei–Gojobori method [] with Jukes–Cantor correction [] by using MEGA v5.0. The transition/transversion bias in the calculation was assumed as R = 1.58 by the ML method. Standard errors were calculated from 1,000 bootstrapping replicates., Ratios of dN/dS were calculated for each of the putative PBRs, non-PBRs, and for the overall exon 2 region. The extent of selective pressure at the amino acid level was measured by the ratio of nonsynonymous to synonymous rates (ω = dN/dS). We estimated ω ratios with the maximum likelihood method by using CODEML in the PAML program version 4.2b []. Several models that differed in their hypotheses concerning the statistical distributions of the ω ratio [,] were considered in this study. Nested models can be compared in pairs by using a likelihood-ratio test []. A Bayesian approach (Bayes empirical Bayes) in CODEML is usually used to predict codons that were potentially under positive selection or balancing selection in models M2a and M8 []. Although it is difficult to distinguish between positive selection and balancing selection, we use this Bayesian approach to estimate sites under balancing selection because MHC genes are thought to be affected by balancing selection rather than positive selection [,]. The ML method has the potential to overestimate ω values because of the existence of recombination []. Therefore, we estimated the ω value by using a Bayesian method with OmegaMap [], which enabled us to infer balancing selection in the presence of recombination. The average number of nonsynonymous substitutions in the PBRs (KB) was also estimated using method II, as described by Satta et al. []., The GENECONV program was used with default settings. The global test for reciprocal recombination or gene conversion events was performed with 10,000 permutations of the sequence alignment to assess significance. Moreover, we examined the possibility of recombination by using the method described by Satta []. In the method, it is assumed that the number of substitutions in a particular region to the entire region is binomially distributed. We divided exon 2 into two […]

Pipeline specifications

Software tools MEGA-V, PAML, omegaMap