Skip to main content

Table 3 Classifier performance using all features for ortholog detection

From: Detection of gene orthology from gene co-expression and protein interaction networks

Datasets

Adaboost j48 AUC

NB AUC

SVM AUC

Log. Reg. AUC

Ensemble AUC

Mouse-Human (PPI)

95.19 (2)

88.72 (5)

90.78 (3)

89.57 (4)

96.18 (1)

Mouse-Human (gene-coexpression)

89.80 (5)

94.1 (4)

97.50 (1)

97.30 (2)

96.10 (3)

Mouse-Fly (PPI)

90.31 (1)

85.81 (3)

81.28 (4)

80.67 (5)

88.94 (2)

Mouse-Yeast (PPI)

92.04 (3)

85.50 (4)

79.63 (5)

95.60 (1)

95.50 (2)

Human-Fly (PPI)

88.18 (1)

83.10 (4)

75.03 (5)

87.04 (3)

87.20 (2)

Human-Yeast (PPI)

82.83 (2)

81.26 (4)

78.22 (5)

81.57 (3)

84.84 (1)

Yeast-Fly (PPI)

74.52 (1)

69.36 (4)

64.57 (5)

74.33 (2)

72.78 (3)

Average Rank (PPI Only)

1.67

4

4.5

3

1.83

Average Rank (PPI+GeneCoexpression)

2.14

4

4

2.86

2

  1. Performance of all the combined features (Reciprocal BLAST hit score, 1 and 2 hop shortest path graph kernel score, 1 and 2 hop random walk graph kernel score, BaryCenter, betweenness, degree distribution and HITS) as input to the decision tree (j48), Naive Bayes (NB), Support Vector Machine (SVM) and Ensemble classifiers on the fly, yeast, human and mouse protein-protein interaction datasets from DIP as well as the gene coexpression networks for mouse and human from GEO. Values in parenthesis are the ranks for the classifiers on the specified dataset.