Submission Details
Molecule(s):
COc1ccc(N(C)C(=O)c2ccccn2)cc1OC
C=C(CC)N(C(=O)c1ccccn1)c1ccc(OC)c(OC)c1
O=C(c1ccccn1)N1CC(c2ccc3c(c2)OCO3)C=N1
COc1ccc(C2CC(C(F)=C(N)N)=NN2C(=O)c2ccccn2)cc1OC
CN(C(=O)c1ccccn1)c1ccc2c(c1)OCO2
COc1ccc(C2CC(C(Cl)=C(N)N)=NN2C(=O)c2ccccn2)cc1OC
COc1ccc(C2CC(F)S(N)=NN2C(=O)Cc2ccccn2)cc1OC
C=C(F)S1=NN(C(=O)c2ccccn2)CC(O)(F)C1c1ccc(SC)c(OC)c1
COc1cc(C2C(F)CN(C(=O)c3ccccn3)N=S2N)ccc1SC
c1ccc(C2CC2c2ccc3c(c2)OCCO3)nc1
COc1ccc(C2CC2C(F)c2ccccn2)cc1OC
COc1ccc(C2CC(C)=NN2C(=O)c2ccccn2)cc1OC
Design Rationale:
Used machine learning to make an adventurous exploration of the chemical space around the most potent molecule from the Covid Moonshot activity data, i.e. MAT-POS-916a2c5a-1. Analysis with SwissADME to submit a physicochemical diverse selection of molecules. Several of the molecules predicted by the model were already in the moonshot database.
Other Notes:
In collaboration with Jeriek van Den Abeele.