Table 3.
Summary of Datasets in PharmaBench: ‘Property Name’ refers to the name of the dataset.
| Category | Property Name | Entries After Data Processing | Final Entries for AI modeling | Unit | Mission Type |
|---|---|---|---|---|---|
| Physochemical | LogD | 14,141 | 13,068 | — | regression |
| Water Solubility | 14,818 | 11,701 | log10nM | regression | |
| Absorption | BBB | 12,486 | 8,301 | — | classification |
| Distribution | PPB | 1,310 | 1,262 | % | regression |
| Metabolism | CYP 2C9 | 4,507 | 999 | Log10uM | regression |
| CYP 2D6 | 1214 | Log10uM | regression | ||
| CYP 3A4 | 1980 | Log10uM | regression | ||
| Clearance | HLMC | 5,252 | 2286 | Log10(mL.min-1.g-1) | regression |
| RLMC | 1129 | Log10(mL.min-1.g-1) | regression | ||
| MLMC | 1403 | Log10(mL.min-1.g-1) | regression | ||
| Toxicity | AMES | 24,780 | 9,139 | — | classification |
| Total | 77,294 | 52,482 |
‘Entries After Data Processing’ indicates the number of entries remaining after the data processing workflow. ‘Final Entries for AI Modeling’ denotes the number of entries used in the final AI modeling process. ‘Unit’ specifies the measurement unit for regression tasks. ‘Mission Type’ encompasses two categories: regression and classification.