Skip to main content

Table 3 Performance of various algorithm models under feature subsets (excluding mitochondrial-specific features)

From: MmisAT and MmisP: an efficient and accurate suite of variant analysis toolkit for primary mitochondrial diseases

Methods

Accuracy (%)

Precision (%)

AUC

F1Score

Recall (%)

MCC

Sub_AdaBoost

80.37

77.87

0.871

0.813

85.53

0.654

Sub_Decision Tree

72.39

75.13

0.724

0.705

66.83

0.654

Sub_Random Forest

81.82

83.20

0.880

0.813

79.65

0.739

Sub_Logistic Regression

80.91

81.05

0.900

0.809

80.93

0.702

Sub_KNeighbors

73.71

70.50

0.802

0.756

81.87

0.601

Sub_SVM

79.21

80.47

0.880

0.787

77.32

0.703

  1. Vari_Train: 1936 benign, 1936 pathogenic
Back to article page

Human Genomics

ISSN: 1479-7364