Free energy perturbation (FEP) is a well-recognized computational technique to predict the relative affinities of protein-ligand interactions, however, the exact accuracy of FEP prediction remains uncertain. Here, the authors assess the current accuracy of the leading FEP workflow by assembling a benchmark dataset, and highlight reliable protocols to maximize the accuracy of FEP in prospective studies.
- Gregory A. Ross
- Chao Lu
- Lingle Wang
