Table 2. Relative method performance based on simulated data for recently diverged species.
| Data set | NJ (liberal) | NJ (strict) | PAR (liberal) | PAR (strict) | NN | BLAST | DNA-BAR | BLOG |
| Ne = 1,000 | 83.69 | 83.58 | 73.31 | 73.14 | 86.18 | 86.18 | 86.25 | 85.96 |
| Ne = 10,000 | 85.53 | 84.27 | 79.79 | 78.38 | 86.11 | 86.09 | 86.83 | 88.15 |
| Ne = 50,000 | 84.20 | 77.35 | 79.53 | 72.32 | 84.76 | 84.56 | 85.24 | 84.58 |
| overall | 84.47a | 81.73b | 77.54c | 74.61d | 85.68e | 85.61 e | 86.11 e | 86.23 e |
DNA barcode query identification success scores (%, N = 100) of six methods applied to barcode sequence datasets simulated under three different effective population sizes (Ne). NJ = neighbor joining, PAR = parsimony, NN = nearest neighbor. Highest scores are in boldface. Overall success scores (%, N = 300) not significantly different in post-hoc pairwise Wilcoxon tests are indicated by same superscripts.