Table 2 Comparison of matrix reductions with empirical data using mare and simple predefined thresholds
From: Selecting informative subsets of sparse supermatrices increases the chance to find correct trees
Data | Reduction | Number of taxa | Resolution score |
|---|---|---|---|
Original | Unreduced | 70 | 91.0896 |
Without 6genome taxa | Unreduced | 64 | 82.1475 |
mare -t 1.67 | 48 | 87.3778 | |
mare with B∗ | Default | 12 | 99.5556 |
-t 3 | 13 | 100 | |
-t 4 | 20 | 94 | |
-t 6 | 22 | 95.8421 | |
-t 7 | 26 | 96.1739 | |
All taxa constraint | 69 | 87.3485 | |
mare with B | Default | 13 | 99.5 |
-t 3 | 15 | 99.8333 | |
-t 4 | 21 | 95.0556 | |
-t 6 | 66 | 88.5238 | |
-t 7 | 67 | 88.4375 | |
All taxa constraint | 69 | 85.803 | |
Simple thresholdsof coverage | Genes 0.4, taxa 0.4 | 22 | 92.5263 |
Taxa 0.66, genes 0.66 | 26 | 82.2174 | |
Genes 0.4 | 59 | 90.1071 | |
Genes 0.66 | 57 | 80.8704 |