Sergey Ovchinnikov @sokrypton.org · Nov 27 1 / 2
I've also looked at the RfaH example. Once again, ColabFold (single sequences input) struggles, and only 1 of the 5 models gets at least the C-terminal motif in alt. conf. For the model_1 that gets it right, if you change the random seed... it sometimes samples the other conf... 😅 (6/8)
0 likes 1 replies
?
Replies
Besides the examples where it's sometimes possible to predict from single sequence. There have been other examples where it was not possible. If it is memorization, why would AF memorize one example and not another? 🤔 twitter.com/GMondaSilva/... (7/8)