Table 2 Comparison of the four automated bioassay neighboring methods by using the depositor-defined (human-curated) assay neighbors as a benchmark.
From: The Text-mining based PubChem Bioassay neighboring analysis
Neighboring method | Recall | Precision | F1 score |
|---|---|---|---|
| Â | Number of common pairs/Number of depositor-specified pairs | Number of common pairs/Number of assay pairs by the method | Harmonic mean(recall, precision) |
1. Activity based | 41% [3694/8802] | 0.5% [4100/671799] | 0.99% |
2. Target similarity based | 46% [4100/8802] | 0.6% [4100/66497] | 1.18% |
3. Biosystems based | 34% [3008/8802] | 0.2% [3008/160356] | 0.39% |
4. Text-mining based | 86% [7530/8802] | 15.0% [7530/50148] | 25.50% |