Table 2.
MutationFinder with sequence validation was applied to abstracts and full articles (abstract + article body) for 70 PDB entries. Results are compared with manually annotated data. True positives (TP), false positives (FP), false negatives (FN), precision (P), recall (R), and F-meaure (F) are presented, describing each approach’s ability to replicate manually curated data
| Metric | Input | TP | FP | FN | P | R | F |
|---|---|---|---|---|---|---|---|
| Normalized Mutations | Abstracts | 1 | 0 | 12 | 1.000 | 0.077 | 0.143 |
| Full text | 10 | 0 | 3 | 1.000 | 0.769 | 0.870 | |
| Mutant Entry Id. | Abstracts | 1 | 0 | 9 | 1.000 | 0.100 | 0.182 |
| Full text | 7 | 0 | 3 | 1.000 | 0.700 | 0.824 |