Open fuxuliu opened 4 months ago
I've recently released a new benchmark that allows users to evaluate some of the most recent structure generation methods for protein-ligand docking (e.g., including RFAA on the PoseBusters Benchmark dataset). You can find code, data, and reproducibility instructions here: https://github.com/BioinfoMachineLearning/PoseBench
Hi, I used rfaa to test the posebuster set, but the precision I got was not as good as the 42% docking success rate mentioned in the article. I think my test protocol is not consistent with yours. Could you please share a complete test script to reproduce this precision?
Thanks.