You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I was wondering if I could use the pre-trained model checkpoint file to assess PPI in few plasmodium proteins that I want to check the interaction in or if the model is not generic?
The text was updated successfully, but these errors were encountered:
While the RAPPPID weights here are indeed trained exclusively on Human proteins, RAPPPID's cross-species performance is shown to surpass leading methods in our latest work.
That work actually outlines INTREPPPID, a model we designed to specifically address the cross-species task. We find INTREPPPID to have superior performance to RAPPPID, especially in the cross-species case.
Note that if there is sufficient PPI data on your organism of interest, you are likely to get the best results training a RAPPPID model directly on that data.
Finally, the easiest way to use the existing human RAPPPID and INTREPPPID weights is via ppi.bio. It's in beta, so there may be some bugs, but generally speaking it's a very fast way to infer PPI using our algorithms.
Hi @jszym
I was wondering if I could use the pre-trained model checkpoint file to assess PPI in few plasmodium proteins that I want to check the interaction in or if the model is not generic?
The text was updated successfully, but these errors were encountered: