Table 3 Examination of the cutoff value of ΔΔE using in-house compound data set
Cutoff value (kcal/mol) | −10.0 | −7.5 | −5.0 | −2.5 | 0 | +2.5 | +5.0 | +7.5 | +10.0 |
|---|---|---|---|---|---|---|---|---|---|
Accuracy | 63.5% | 62.4% | 62.4% | 65.9% | 69.4% | 70.6% | 64.7% | 64.7% | 63.5% |
Sensitivity | 85.7% | 82.8% | 82.8% | 77.5% | 74.5% | 75.0% | 68.4% | 67.8% | 66.7% |
Specificity | 52.6% | 51.8% | 51.8% | 55.6% | 61.8% | 63.6% | 57.1% | 57.7% | 56.0% |
Positive prediction value | 47.1% | 47.1% | 47.1% | 60.8% | 74.5% | 76.5% | 76.5% | 78.4% | 78.4% |
Negative prediction value | 88.2% | 85.3% | 85.3% | 73.5% | 61.8% | 61.8% | 47.1% | 44.1% | 41.2% |
Matthews correlation coefficient | −0.2 | −0.2 | −0.2 | −0.2 | 0.4 | 0.4 | 0.2 | 0.2 | 0.2 |
Coverage | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% |
Inconclusive | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
Out of domain | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |