Skip to main content

Table 8 Comparisons of different models for hot spots prediction on 10-fold cross-validation test

From: Rigorous assessment and integration of the sequence and structure based features to predict hot spots

Method P R F1 ΔF1 P-value
Random 0.56 0.47 0.48 -- --
Sequence-based SVM 0.65 0.65 0.65 +0.17 0.003
Structure-based SVM 0.65 0.60 0.61 +0.13 0.02
All features SVM 0.66 0.68 0.66 +0.18 0.002