TABLE 1.
Summary information about the benchmarking dataset is shown. The columns labeled “Tc >X” show the percentage of active Tanimoto similarities above cutoffs 0.8 and 0.3 reported by Jasial et al.36 The meanSIM columns shows average recovery rates reported by Heikamp et al.35
| Difficulty Class (# Targets) | # of Compounds Median [Range] | MACCS | ECFP 4 | ||
|---|---|---|---|---|---|
| Tc >0.8 | meanSim | Tc >0.3 | meanSim | ||
| Easy (22) | 116 [50–339] | 16.0% | 74.1% | 38.2% | 89.3% |
| Intermediate (50) | 371 [111–1540] | 5.0% | 42.6% | 10.9% | 69.4% |
| Difficult (30) | 1609 [420–3144] | 1.2% | 13.8% | 4.1% | 27.8% |