Skip to main content

Table 1 Relative performance of CONCATERPILLAR and Clusterflock

From: Clusterflock: a flocking algorithm for isolating congruent phylogenomic datasets

 

0 %

10 %

20 %

30 %

2

0.91/0.84

0.99/0.87

0.94/0.90

0.90/0.94

5

1.00/0.65

1.00/0.67

0.93/0.68

0.75/0.66

10

1.00/0.54

1.00/0.56

0.89/0.56

0.47/0.44

15

1.00/0.43

1.00/0.46

0.83/0.47

0.38/0.25

20

1.00/0.37

0.99/0.41

0.72/0.37

0.22/0.19

25

1.00/0.29

0.99/0.35

0.69/0.34

0.24/0.16

  1. We tabulated Jaccard Indices of CONCATERPILLAR runs on the simulated datasets described at 0, 10, 20, and 30 % missing data (columns) across 2, 5, 10, 15, 20, and 25 simulated topologies (rows). For comparison, the average CONCATERPILLAR JI is shown alongside the average Clusterflock JI in each cell (CONCATERPILLAR|clusterflock). At 40 and 50 % missing data, CONCATERPILLAR failed to run (denoted by “NA”)