search

vsomnath / graphretro

Learning Graph Models for Retrosynthesis Prediction (NeurIPS 2021)
https://arxiv.org/abs/2006.07038
MIT License
46 stars 14 forks source link

multi edit product problem #9

Closed queliyong closed 1 year ago

queliyong commented 1 year ago

Hi, Considering this multi edit product and it's corresponding reactants: C1=CC=C(N([H])C2=CC=CC=C2)C=C1.C1=CC=C(N([H])C2=CC=CC=C2)C=C1.C(Br)1=C([H])C([H])=C(C2=C([H])C([H])=C(Br)C([H])=C2[H])C([H])=C1[H]>>C1=CC=C(N(C2=CC=C(C3=CC=C(N(C4C=CC=CC=4)C4=CC=CC=C4)C=C3)C=C2)C2C=CC=CC=2)C=C1 A656EFFC-2CED-4b5b-95AE-65A40DA7DDA7 The code and models this repo provided can not generate the reactants above, what can I do to generate this reactants. Thanks very much for your advice.

vsomnath commented 1 year ago

Hi, I think you could try either RetroXpert or any of the transformer models potentially.