Submission Details
Molecule(s):
CC(=O)N1CCN(C(=O)C(F)c2ccccn2)CC1
NC(=O)N1CCN(C(=O)C(O)(F)c2ccccn2)CC1
CC(=O)N1CCN(C(=O)Cc2ccccn2)CC1
CC(=O)N1CCN(C(=O)C(F)c2ncccc2Cl)CC1
CC(=O)N1CCC(S(=O)(=O)c2ccccc2Cl)CC1
CC(=O)N1CCC(C(F)C(C#N)c2cccc(F)c2S(N)(=O)=O)CC1
C[C@H](NC(=O)[C@@H](C)c1ccc(F)c(F)c1)c1cccc(F)c1
N#Cc1c(F)cccc1C(F)C(=O)N1CCN(C(N)=O)CC1
Cc1ccccc1CN1CCN(C(=O)C(F)C(C)C)CC1
CCC(=O)N1CCN(Cc2cc(C#N)ccc2C(C)(O)F)CC1
CCC(=O)N1CCN(Cc2ccccc2CN2CCN(C(=O)C(O)F)CC2)CC1
N#CC(C1C=CC(F)=C1)C1CCN(C(CO)c2cccc(F)c2)CC1
CC(=O)NC(c1cc(F)cc(S(N)(=O)=O)c1)C(C)F
C[C@H](NC(=O)C(F)F)c1cccc(F)c1
CCC(=O)NCc1ccccc1CN1CCN(C(=O)CN2CCN(C(=O)C(F)CC)CC2)CC1
Cc1ccccc1CN1CCN(C(=O)N2CCN(C(=O)C(C)F)CC2)CC1
Design Rationale:
Design rationale: Started out from a very broad collection of known anti-viral compounds (HIV-1 protease inhibitors, Ebola drugs, antivirals used to treat herpes, ...) and related structures. Then applied machine learning to this data-set to learn the chemical space of anti-virals and used this model to generate a diverse set of compounds containing partial similarities (shape and properties) to fragment crystal structures X_0689, X_0769, X_0831, X_1382. The final selection of 16 molecules (4 per fragment) was based on a range of factors like synthesis accessibility, molecular weight, diversity, lipophilicity and lack of toxicity.
Other Notes:
Special thanks to Jeriek Van den Abeele at UiO's Department of Physics for computational support.