Evaluation of the utility of data-dependent molecular property associations applied to enhance preclinical predictivity of human CL
Data-Dependent Prediction Method | Number of Predicted Inlier/Outlier/Not Predicted | Total Number (%) of Correct Predictions | Number (%) of Predicted Inliers Correctly Predicted | Number (%) of Predicted Outliers Correctly Predicted | Overall MAE (%) | MAE for Predicted Inliers (%) | MAE for Predicted Outliers (%) |
|---|---|---|---|---|---|---|---|
| ml/min/kg | |||||||
| Qualitative (Class) Rat | 3/12/6 | 12 (80.0) | 2 (66.7) | 10 (83.3) | 8.38 (113) | 3.14 (93.2) | 10.91 (118) |
| Qualitative (Class) Dog | 5/11/5 | 13 (81.3) | 4 (80.0) | 9 (81.8) | 11.8 (159) | 2.63 (81.9) | 15.6 (163) |
| Qualitative (Class) Monkey | 15/0/6 | 7 (46.7) | 7 (46.7) | N.A. | 6.08 (81.9) | 5.79 (93.7) | N.A. |
| Quantitative (2-fold) Rat | 6/12/3 | 10 (55.6) | 4 (66.7) | 6 (50.0) | 8.38 (113) | 5.92 (102) | 10.91 (118) |
| Quantitative (2-fold) Dog | 0/0/21 | N.A. | N.A. | N.A. | 11.8 (159) | N.A. | N.A. |
| Quantitative (2-fold) Monkey | 18/0/3 | 9 (50.0) | 9 (50.0) | N.A. | 6.08 (81.9) | 5.90 (83.6) | N.A. |
N.A., not applicable.