Submission Details
Molecule(s):
O=C(N[C@@H](CC1CCCCC1)C(=O)N[C@@H](C[C@@H]1CCNC1=O)C(O)S(=O)(=O)O)OCc1cccc(Br)c1
O=C(N[C@@H](CC1CCCCC1)C(=O)N[C@@H](C[C@@H]1CCNC1=O)C(O)S(=O)(=O)O)OCc1cccc(Cl)c1
CC(C)C[C@H](NC(=O)OCc1cccc(F)c1)C(=O)N[C@@H](CC1CCNC1=O)C(O)S(=O)(=O)O
CC(C)C[C@H](NC(=O)OCCC1CCCCC1)C(=O)N[C@@H](CC1CCNC1=O)C(O)S(=O)(=O)O
Design Rationale:
Built a moderately-well performing machine learning algorithm on the molecules in the activity data from this moonshot project. Ran this model on 1.9 million molecules from ChEMBL and predicted their pIC50 values. The four most active molecules are shown here.