Research Article
Performance Comparison between the Specific and Baseline Prediction Models of Ecotoxicity for Pharmaceuticals: Is a Specific QSAR Model Inevitable?
Table 1
Summary statistics for performance comparison of Sangion’s model versus ECOSAR.
| Measure of predictive ability | Sangion’s model | ECOSAR | Whole dataset | In AD |
| Number of pharmaceuticals | 41 | 36 | 41 | Pearson correlation coefficient | 0.422 | 0.483 | 0.272 | Root mean square error, RMSE (log units) | 1.630 | 1.422 | 2.744 | Mean absolute residual (log units) | 1.190 | 1.008 | 1.825 | Minimum absolute residual (log units) | 0.025 | 0.025 | 0.084 | Maximum absolute residual (log units) | 5.023 | 5.023 | 9.858 | % pharmaceuticals <factor of 2 | 12.2 | 13.9 | 19.5 | % pharmaceuticals <factor of 5 | 36.6 | 41.7 | 34.1 | % pharmaceuticals <factor of 10 | 51.2 | 58.3 | 48.8 | % pharmaceuticals <factor of 100 | 85.4 | 94.4 | 61.0 | % pharmaceuticals <factor of 1000 | 92.7 | 94.4 | 85.4 |
|
|