Submission Details
Molecule(s):
COC1C=C(O)CC1NCCN1C(C)=CCC(C)=C1C(O)/C=C/F
C=CC1=C(NCC(=O)NC2C=NC=CC2)C(C)C(C(C)(C)C)=C1O
CC(C)(C)C1=C(O)C(CC#N)=C(N2CCC(O)CC2)C1OC(C)(F)C(N)=O
C=C1N2C(NCS(N)(=O)=O)=C(C)CCC2CCN1[C@H]1CC[C@@H](C(=O)NC)O1
CC(O)N1CCN(C2=C(C#N)C(F)=C(Cl)CC2C(O)S(N)(=O)=O)CC1C
CCNC1CC=C(F)C=C1CNC(=O)[C@@]1(C)[C@H]2C(CC[C@@]2(C)O)C2(C)CCC21C(=O)O
CNC(=O)C1=C(C)CCC2C(CO)CN(CCS(N)(=O)=O)CN12
CC1CCCC(NCS(N)(=O)=O)N1CC/C(F)=C/CO
Design Rationale:
These designs are based on a generative machine learning model trained on very broad collection of known anti-viral compounds (HIV-1 Protease inhibitors, Ebola and Herpes antivirals, ...) and a hand curated selection of related molecules. The set of molecules generated by this model are then scored on chemical similarity to the COVID-19 / SARS-CoV-2 protease transition state analogue inhibitor template generated by the Quantum Corona project. The resulting high-scoring molecules were filtered on a series of AMDE and toxicity conditions, including Veber, Egan and Muegge violations. All molecules presented here are P-glycoprotein substrates, have a high GI absorption and have a favorable bioavailability score. The final subsection of molecules presented here was decided on by hand based on a range of factors including synthetic accessibility and some diversity measures.
Other Notes:
Special thanks to Jeriek Van den Abeele at UiO's Department of Physics for computational support and to Jarvist Moore Frost at Imperial College London's Blackett Laboratory for kick-starting the the Quantum Corona project and making the protease transition state analogue inhibitor template widely available.