TDC.DRD3 Leaderboard
Docking Target Summary
| Dataset | Diseases | Link to DRD3 target protein |
|---|---|---|
TDC.DRD3 |
Tremor, Schizophrenia | Uniprot Page |
Leaderboard: Models Make at Most 5000 Oracle Calls
| Rank | Model | Contact | Link | #Params | Top100 | Top10 | Top1 | Diver. | Novel. | %Pass | Top 1-P | M1 | Molecules |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| - | Best-in-data | - | - | - | -12.080 | -12.590 | -12.800 | 0.864 | - | 0.780 | -11.700 | 5.100 | Link |
| 1 | Graph-GA | Tianfan Fu | GitHub, Paper | 0 | -14.811 ± 0.413 | -15.930 ± 0.336 | -16.533 ± 0.309 | 0.626 ± 0.092 | 1.000 ± 0.000 | 0.393 ± 0.308 | -14.267 ± 0.450 | 9.669 ± 0.468 | Link |
| 2 | SMILES-LSTM | Tianfan Fu | GitHub, Paper | 3,149,000 | -13.017 ± 0.385 | -14.030 ± 0.421 | -14.533 ± 0.525 | 0.740 ± 0.056 | 1.000 ± 0.000 | 0.257 ± 0.103 | -12.533 ± 0.403 | 5.826 ± 1.908 | Link |
| 3 | GCPN | Tianfan Fu | GitHub, Paper | 17,600 | -10.045 ± 0.226 | -11.483 ± 0.581 | -12.300 ± 0.993 | 0.922 ± 0.002 | 1.000 ± 0.000 | 0.167 ± 0.045 | -9.367 ± 0.170 | 10.000 ± 0.000 | Link |
| 4 | MARS | Tianfan Fu | GitHub, Paper | 153,000 | -9.509 ± 0.035 | -10.693 ± 0.172 | -11.433 ± 0.450 | 0.873 ± 0.002 | 1.000 ± 0.000 | 0.527 ± 0.087 | -9.000 ± 0.082 | 7.073 ± 0.798 | Link |
| 5 | MolDQN | Tianfan Fu | GitHub, Paper | 2,694,800 | -8.236 ± 0.089 | -9.348 ± 0.188 | -9.990 ± 0.194 | 0.893 ± 0.005 | 1.000 ± 0.000 | 0.023 ± 0.012 | -7.980 ± 0.112 | 10.000 ± 0.000 | Link |
: The lower/higher, the better. Click to sort based on the specific column.
* Rank is based on Top 100 average docking scores.
** Novelty is calculated against the ZINC training dataset.
Leaderboard: Models Make at Most 1000 Oracle Calls
| Rank | Model | Contact | Link | #Params | Top100 | Top10 | Top1 | Diver. | Novel. | %Pass | Top 1-P | M1 | Molecules |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| - | Best-in-data | - | - | - | -12.080 | -12.590 | -12.800 | 0.864 | - | 0.780 | -11.700 | 5.100 | Link |
| 1 | Graph-GA | Tianfan Fu | GitHub, Paper | 0 | -11.224 ± 0.484 | -12.400 ± 0.782 | -13.233 ± 0.713 | 0.815 ± 0.046 | 1.000 ± 0.000 | 0.777 ± 0.096 | -10.600 ± 0.374 | 7.695 ± 0.909 | Link |
| 2 | SMILES-LSTM | Tianfan Fu | GitHub, Paper | 3,149,000 | -9.971 ± 0.115 | -11.163 ± 0.141 | -11.967 ± 0.205 | 0.871 ± 0.004 | 1.000 ± 0.000 | 0.777 ± 0.026 | -9.367 ± 0.094 | 4.818 ± 0.541 | Link |
| 3 | GCPN | Tianfan Fu | GitHub, Paper | 17,600 | -9.053 ± 0.080 | -11.027 ± 0.273 | -12.033 ± 0.618 | 0.913 ± 0.001 | 1.000 ± 0.000 | 0.170 ± 0.022 | -8.167 ± 0.047 | 10.000 ± 0.000 | Link |
| 4 | MARS | Tianfan Fu | GitHub, Paper | 153,000 | -8.224 ± 0.196 | -9.843 ± 0.068 | -11.100 ± 0.141 | 0.871 ± 0.004 | 1.000 ± 0.000 | 0.563 ± 0.052 | -7.367 ± 0.205 | 6.037 ± 0.137 | Link |
| 5 | MolDQN | Tianfan Fu | GitHub, Paper | 2,694,800 | -6.738 ± 0.042 | -7.506 ± 0.085 | -7.800 ± 0.042 | 0.904 ± 0.001 | 1.000 ± 0.000 | 0.033 ± 0.005 | -6.450 ± 0.085 | 10.000 ± 0.000 | Link |
: The lower/higher, the better. Click to sort based on the specific column.
* Rank is based on Top 100 average docking scores.
** Novelty is calculated against the ZINC training dataset.
Leaderboard: Models Make at Most 500 Oracle Calls
| Rank | Model | Contact | Link | #Params | Top100 | Top10 | Top1 | Diver. | Novel. | %Pass | Top 1-P | M1 | Molecules |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| - | Best-in-data | - | - | - | -12.080 | -12.590 | -12.800 | 0.864 | - | 0.780 | -11.700 | 5.100 | Link |
| 1 | Graph-GA | Tianfan Fu | GitHub, Paper | 0 | -10.036 ± 0.221 | -11.527 ± 0.533 | -12.500 ± 0.748 | 0.857 ± 0.005 | 1.000 ± 0.000 | 0.710 ± 0.080 | -9.300 ± 0.163 | 6.493 ± 0.341 | Link |
| 2 | SMILES-LSTM | Tianfan Fu | GitHub, Paper | 3,149,000 | -9.419 ± 0.173 | -10.687 ± 0.335 | -11.367 ± 0.579 | 0.875 ± 0.005 | 1.000 ± 0.000 | 0.727 ± 0.012 | -8.767 ± 0.170 | 5.787 ± 0.934 | Link |
| 3 | GCPN | Tianfan Fu | GitHub, Paper | 17,600 | -8.119 ± 0.104 | -10.230 ± 0.354 | -11.967 ± 0.680 | 0.914 ± 0.001 | 1.000 ± 0.000 | 0.127 ± 0.005 | -7.200 ± 0.141 | 10.000 ± 0.000 | Link |
| 4 | MARS | Tianfan Fu | GitHub, Paper | 153,000 | -7.278 ± 0.198 | -9.067 ± 0.377 | -9.833 ± 0.309 | 0.866 ± 0.005 | 1.000 ± 0.000 | 0.660 ± 0.050 | -6.100 ± 0.141 | 5.827 ± 0.937 | Link |
| 5 | MolDQN | Tianfan Fu | GitHub, Paper | 2,694,800 | -6.357 ± 0.084 | -7.173 ± 0.166 | -7.620 ± 0.185 | 0.903 ± 0.002 | 1.000 ± 0.000 | 0.030 ± 0.016 | -6.030 ± 0.073 | 10.000 ± 0.000 | Link |
: The lower/higher, the better. Click to sort based on the specific column.
* Rank is based on Top 100 average docking scores.
** Novelty is calculated against the ZINC training dataset.
Leaderboard: Models Make at Most 100 Oracle Calls
| Rank | Model | Contact | Link | #Params | Top100 | Top10 | Top1 | Diver. | Novel. | %Pass | Top 1-P | M1 | Molecules |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| - | Best-in-data | - | - | - | -12.080 | -12.590 | -12.800 | 0.864 | - | 0.780 | -11.700 | 5.100 | Link |
| 1 | SMILES-LSTM | Tianfan Fu | GitHub, Paper | 3,149,000 | -7.594 ± 0.182 | -10.033 ± 0.186 | -11.133 ± 0.634 | 0.884 ± 0.002 | 1.000 ± 0.000 | 0.763 ± 0.019 | -1.100 ± 1.417 | 5.219 ± 0.247 | Link |
| 2 | Graph-GA | Tianfan Fu | GitHub, Paper | 0 | -7.222 ± 0.013 | -10.177 ± 0.158 | -11.767 ± 1.087 | 0.885 ± 0.001 | 1.000 ± 0.000 | 0.693 ± 0.037 | 0.000 ± 0.000 | 5.223 ± 0.256 | Link |
| 3 | MARS | Tianfan Fu | GitHub, Paper | 153,000 | -5.928 ± 0.298 | -8.133 ± 0.328 | -9.100 ± 0.712 | 0.873 ± 0.010 | 1.000 ± 0.000 | 0.807 ± 0.033 | -3.633 ± 0.946 | 4.470 ± 1.047 | Link |
| 4 | MolDQN | Tianfan Fu | GitHub, Paper | 2,694,800 | -5.178 ± 0.341 | -6.438 ± 0.176 | -7.020 ± 0.194 | 0.907 ± 0.001 | 1.000 ± 0.000 | 0.017 ± 0.012 | -3.630 ± 2.588 | 10.000 ± 0.000 | Link |
| 5 | GCPN | Tianfan Fu | GitHub, Paper | 17,600 | 3.860 ± 0.102 | -5.617 ± 0.413 | -11.633 ± 2.217 | 0.909 ± 0.001 | 1.000 ± 0.000 | 0.093 ± 0.009 | 7.667 ± 0.262 | 10.000 ± 0.000 | Link |
: The lower/higher, the better. Click to sort based on the specific column.
* Rank is based on Top 100 average docking scores.
** Novelty is calculated against the ZINC training dataset.