Table 1 Number of proteins and functions in BP, CC and MF ontologies in the dataset
From: TEMPROT: protein function annotation using transformers embeddings and homology search
| Â | BP | CC | MF |
|---|---|---|---|
Training set | 47,691 | 45,309 | 32,421 |
Validation set | 5252 | 4985 | 3587 |
Test set | 2392 | 1265 | 1137 |
Functions | 3992 | 551 | 677 |