Skip to main content

Table 4 Estimates of AUC-ROC and AUC-PR of different methods for balanced dataset in predicting donor splice sites using human, bovine, fish and worm species

From: Identification of donor splice sites using support vector machine: a computational approach based on positional, compositional and dependency features

 

Species

Approaches

MM1-SVM

LIK-SVM

WD-SVM

WDS-SVM

EFFECT

Proposed

AUC-ROC

Human

97.07

97.13

97.25

97.06

97.15

96.05

Bovine

96.98

97.63

97.83

97.59

97.70

96.94

Fish

97.24

97.34

97.68

97.53

97.59

96.95

Worm

97.49

98.02

98.23

98.12

98.15

96.24

AUC-PR

Human

96.78

97.52

97.67

97.38

97.58

97.64

Bovine

96.66

97.48

97.59

97.26

97.51

97.89

Fish

96.85

97.42

97.67

97.39

97.49

97.91

Worm

96.92

97.51

97.78

97.63

97.71

97.90

  1. It can be seen that the values of AUC-ROC of the proposed approach are less as compared to that of others, whereas the values of AUC-PR for the proposed approach are at par with that of other approaches (except MM1-SVM), in all the four species