< Back to Article

Inferring Phylogenetic Networks with Maximum Pseudolikelihood under Incomplete Lineage Sorting

Fig 6

Performance (average computing time per replicate) of SNaQ and PhyloNet.

in simulations using true gene trees on networks with n = 6, 10 or 15 taxa and h = 1, 2 or 3. Each replicate consisted of 10 independent runs with full optimization of branch lengths and inheritance probabilities for each run. Pie charts display accuracy (black: probability of recovering the true network). With n = 10 and 300 or more loci, or with n = 15, PhyloNet was too slow to run.

Fig 6