Skip to main content

Table 2 The number of proteins and the number of annotations in the train and test sets with respect to the three setups for human

From: A close look at protein function prediction evaluation protocols

  

Training set

Test set

 

Set.

Proteins

Annots.

Proteins

Annots.

F

CV

4532

8467

1133

2116

 

NA

4305

6898

799

1343

 

NP

4305

6898

1344

2174

P

CV

7533

31794

1883

7948

 

NA

5824

12196

3301

13192

 

NP

5824

12196

3574

12973

C

CV

8440

19196

2110

4799

 

NA

5082

8185

2966

5511

 

NP

5082

8185

5468

10200

  1. F, P and C represent molecular function, biological process and cellular component, respectively. For the CV setup, numbers represent average values computed across the training and test folds (5-fold CV)