Table 3.
Comparison of predictive performance on COACH420 and HOLO4K datasets
| COACH420 | HOLO4K | |||
|---|---|---|---|---|
| Top-n | Top-(n+2) | Top-n | Top-(n+2) | |
| Fpocket | 56.4 | 68.9 | 52.4 | 63.1 |
| Fpocket+PRANKa | 63.6 | 76.5 | 62.0 | 71.0 |
| SiteHound† | 53.0 | 69.3 | 50.1 | 62.1 |
| MetaPocket 2.0† | 63.4 | 74.6 | 57.9 | 68.6 |
| DeepSite† | 56.4 | 63.4 | 45.6 | 48.2 |
| P2Rank[protrusion]b | 64.2 | 73.0 | 59.3 | 67.7 |
| P2Rank | 72.0 | 78.3 | 68.6 | 74.0 |
The numbers represent identification success rate [%] measured by DCCcriterion (distance from pocket center to closest ligand atom) with 4 Å threshold considering only pockets ranked at the top of the list (n is the number of ligands in considered structure)
†These methods failed to produce predictions for some portion of input proteins. Here we display success rates calculated only based on subsets of proteins, on which they finished successfully. Detailed, pairwise comparison with P2Rank on the exact subsets can be found in the Additional file 1.
aPredictions of Fpocket re-scored by PRANK algorithm (which is included in P2Rank software package)
bReduced version of P2Rank that uses only single geometric feature: protrusion