TABLE 4.
Evaluation of classifiers to predict the presence or absence of species across the 140 mock communities with and without optimizing the F1 score (mean ± std. dev.).
| Classifier | Precision | Recall | F1 score | Precision (F1 optimized) | Recall (F1 optimized) | Optimized F1 score |
| MCP | 0.98 ± 0.04 | 0.94 ± 0.06 | 0.96 ± 0.05 | – | – | – |
| Unfiltered MCP | 0.88 ± 0.13 | 0.97 ± 0.04 | 0.92 ± 0.09 | 0.94 ± 0.07 | 0.95 ± 0.06 | 0.94 ± 0.06 |
| MCP w/ MGDB | 0.99 ± 0.02 | 0.95 ± 0.04 | 0.97 ± 0.03 | – | – | – |
| Ganon | 0.81 ± 0.20 | 0.99 ± 0.01 | 0.87 ± 0.14 | 0.91 ± 0.10 | 0.94 ± 0.07 | 0.92 ± 0.08 |
| Kraken | 0.69 ± 0.24 | 0.99 ± 0.01 | 0.79 ± 0.18 | 0.87 ± 0.11 | 0.91 ± 0.09 | 0.88 ± 0.09 |
| Bracken | 0.68 ± 0.24 | 1.00 ± 0.01 | 0.78 ± 0.19 | 0.87 ± 0.11 | 0.90 ± 0.10 | 0.88 ± 0.09 |
| MetaCache | 0.80 ± 0.18 | 0.99 ± 0.01 | 0.88 ± 0.12 | 0.90 ± 0.09 | 0.95 ± 0.05 | 0.92 ± 0.07 |
| Centrifuge | 0.82 ± 0.21 | 0.95 ± 0.05 | 0.86 ± 0.15 | 0.89 ± 0.13 | 0.90 ± 0.10 | 0.90 ± 0.11 |
| DIAMOND-LCA | 0.83 ± 0.15 | 0.72 ± 0.11 | 0.76 ± 0.09 | 0.87 ± 0.11 | 0.70 ± 0.10 | 0.77 ± 0.08 |
| Kaiju | 0.73 ± 0.23 | 0.95 ± 0.04 | 0.80 ± 0.17 | 0.87 ± 0.12 | 0.87 ± 0.09 | 0.87 ± 0.10 |
| mOTUs | 0.90 ± 0.11 | 0.92 ± 0.03 | 0.91 ± 0.07 | 0.91 ± 0.10 | 0.92 ± 0.04 | 0.91 ± 0.07 |
| MetaPhlAn | 0.92 ± 0.03 | 0.73 ± 0.08 | 0.81 ± 0.07 | 0.92 ± 0.03 | 0.73 ± 0.08 | 0.81 ± 0.05 |
MCP and MCP w/MGDB were run with default settings without F1 optimization.