Figure 7
From: BBCA: Improving the scalability of *BEAST using random binning
Average species tree estimation error on one Laurasiatheria 1000bp simulated dataset using BBCA, *BEAST and concatenation with maximum likelihood (CA-ML). BBCA is run with a 24-hour or 48-hour time limit on each 25-gene bin, and *BEAST is run with a 48-hour, 96-hour or 168-hour time limit. BBCA using either 24 hours or 48 hours per bin recovers the true species tree, but both *BEAST analyses and the CA-ML fail to recover the true species tree.