Table 1 Molecular descriptor statistics for the QM9 and AICures molecules in the GEOM dataset.
From: GEOM, energy-annotated molecular conformations for property prediction and molecular generation
AICures drug dataset (N = 304,466) | |||
|---|---|---|---|
Mean | Standard deviation | Maximum | |
Number of atoms | 44.4 | 11.3 | 181 |
Number of heavy atoms | 24.9 | 5.7 | 91 |
Molecular weight (amu) | 355.4 | 80.4 | 1549.7 |
Number of rotatable bonds | 6.5 | 3.0 | 53 |
Stereochemistry (specified) | 45,712 | — | — |
Stereochemistry (all) | 83,326 | — | — |
QM9 dataset (N = 133,258) | |||
Mean | Standard deviation | Maximum | |
Number of atoms | 18.0 | 3.0 | 29 |
Number of heavy atoms | 8.8 | 0.51 | 9 |
Molecular weight (amu) | 122.7 | 7.6 | 152.0 |
Number of rotatable bonds | 2.2 | 1.6 | 8 |
Stereochemistry (specified) | 95,734 | — | — |
Stereochemistry (all) | 95,734 | — | — |
