Too many clash in final designed Binder-Target structure? #112

After running the entire pipeline and obtaining the final structure predicted by AF2, I noticed that in the models with PAE < 10, nearly 80% of the binders exhibited clashes — their amino acids overlapped with those of the target, with some atoms frequently occupying the same space. I’d like to ask: is this level of clash rate considered normal? Do such binders still have a chance of successfully binding in wet-lab experiments?