Research Article

RF-Phos: A Novel General Phosphorylation Site Prediction Tool Based on Random Forest

Table 4

Feature types and their count percentage in the top-ten important features for each phosphosite.

ResiduesFeatures

SASA (30%), CDT (20%), QSO (20%) ACH (10%), OP (10%), and SF (10%)
TCTD (50%), OP (20%), QSO (20%), and SF (10%)
YCDT (60%), ASA (20%), IG (10%), and OP (10%)