Table 1 Descriptors used for the dataset with their initial and final dimensions after feature engineering
Name | Type | Initial length | Final length |
|---|---|---|---|
ECFP1024_4 | Fingerprint | 1024 | 105 |
ECFP1024_6 | Fingerprint | 1024 | 128 |
ECFP2048_4 | Fingerprint | 2048 | 85 |
ECFP2048_6 | Fingerprint | 2048 | 93 |
ECFP4096_4 | Fingerprint | 4096 | 80 |
ECFP4096_6 | Fingerprint | 4096 | 82 |
CATS2D | Categorical | 150 | 118 |
2DAtomPairs | Categorical | 1596 | 387 |
FGroup | Categorical | 153 | 35 |
MACCS | Fingerprint | 167 | 86 |
CATS2D + 2DAtomPairs | Combined | 1746 | 503 |
CATS2D + 2DAtomPairs + ECFP4096_4 | Combined | 5842 | 682 |
FGroup + CATS2D | Combined | 304 | 151 |
FGroup + ECFP1024_6 | Combined | 1178 | 165 |
2DAtomPairs + ECFP4096_4 | Combined | 5693 | 549 |