The optimal power allocation is computed using the bisection method (relative error below 10−5).
In addition, we used the bisection method to find a separate posterior probability cutoff for each of the 16 TFs that gave the SBaSeTraM method a FPR (for that TF) close to the FPR obtained with GMATIM.
Thus, we can use the bisection method to find the optimal global multiplier μ [19].
We use the bisection method to solve this constrained optimization problem.
Here, we used the bisection method (Burden and Faires 1985), but any other simple root-finding algorithm might be used.
In the latter, starting trees were obtained using random sequence addition and swapped using the tree bisection reconnection algorithm.
The ML tree was reconstructed with PAUP* 4 [ 54] under the TrN + Γ + I model of sequence evolution and using the tree bisection reconnection (TBR) branch-swapping algorithm and 100 random sequence addition starting trees.
Conditions for the ML analysis included starting trees obtained by stepwise addition, and branch swapping using the tree-bisection-reconnection (TBR) algorithm.
Maximum parsimony analysis of the combined data set was performed with Paup* v. 4.0 [ 64] using an equally weighted character matrix, heuristic searches using the tree-bisection-reconnection branch-swapping algorithm, and 100 random addition replicates.
One thousand heuristic search replicates were done, using the tree bisection-reconnection branch-swapping algorithm and a random addition of taxa for each replicate, saving 100 trees per replicate.
For this analysis, a simple heuristic search was performed in TNT [ 39, 40] using the tree-bisection-reconnection swapping algorithm with thousands of random addition sequence replicates and 10 trees saved per replicate.
