The probability that trans-specificity is apparent using phylogenetic methods

ρ =
Finite sites model
θ = 0.10.7960.3690.224NANA
θ = 10.8320.4140.1640.0170.002
θ = 100.8330.3970.0810.0000.000
Finite synonymous sites model
θ = 0.10.8010.4000.194NANA
θ = 10.8390.3650.1480.0080.001
θ = 100.8400.2640.0260.0000.000
  • Power was estimated using 1000 replicates for each parameter combination. The selected polymorphism was located in the center of the region, and mutation between allelic classes occurred with symmetric rate 0.01. Other parameters used were x = 0.5, τ = 8, and n = 4 (cf. Table 2). Neutral mutations were added according to a simple Jukes-Cantor substitution model (Jukes and Cantor 1969) using a method similar to the one used by Schierup and Hein (2000). The simulated region was assumed to correspond to 1 kb, but in the “finite synonymous site” model, only one-third of all sites were allowed to vary. Mutations were added according to the rates in the table, but data sets with less than three segregating sites were not used as they provide too little information for phylogenetic methods. We used PHYLIP for phylogenetic reconstruction. Each data set was run through seqboot (to generate 1000 bootstrap data sets), dnadist, neighbor, and finally consense to generate a consensus tree. If the final tree showed a trans-specific topology with a bootstrap support of >70% for the terminal branches, the data set was said to support trans-specificity.