Skip to main content

Table 3 Comparison with state-of-the-art methods on LGG and KIPAN datasets

From: MOCAT: multi-omics integration with auxiliary classifiers enhanced autoencoder

Method

LGG (2 Categories)

KIPAN (3 Categories)

ACC(%)

F1(%)

AUC(%)

ACC(%)

F1\(\_\)w(%)

F1\(\_\)m(%)

(95% CI)

(95% CI)

(95% CI)

(95% CI)

(95% CI)

(95% CI)

KNN

72.9 (68.7-77.1)

73.8 (69.7-77.9)

79.9 (75.2-84.6)

96.7 (95.3-98.1)

96.7 (95.3-98.1)

96.0 (94.3-97.7)

SVM

75.4 (69.7-81.1)

75.7 (69.5-81.9)

75.4 (69.7-81.1)

99.5 (99.1-99.9)

99.5 (99.1-99.9)

99.4 (98.9-99.9)

Lasso

76.1 (73.9-78.3)

76.7 (74.0-79.4)

82.3 (78.9-85.7)

97.4 (97.2-97.6)

97.4 (97.2-97.6)

97.2 (96.7-97.7)

RF

74.8 (73.3-76.3)

74.2 (73.0-75.4)

82.3 (81.1-83.5)

98.1 (97.4-98.8)

98.1 (97.4-98.8)

97.5 (96.1-98.9)

XGBoost

75.6 (70.6-80.6)

76.7 (72.7-80.7)

84.0 (81.1-86.9)

99.3 (98.3-100)

99.3 (98.3-100)

98.9 (97.2-100)

NN

73.7 (70.8-76.6)

74.8 (71.8-77.8)

81.0 (76.4-85.6)

99.1 (98.5-99.7)

99.1 (98.5-99.7)

99.1 (98.5-99.7)

GRridge

74.6 (69.9-79.3)

75.6 (71.1-80.1)

82.6 (77.1-88.1)

99.4 (98.9-99.9)

99.4 (98.9-99.9)

99.3 (98.8-99.8)

BPLSDA

75.9 (72.8-79.0)

73.8 (70.0-77.6)

82.5 (79.6-85.4)

93.3 (91.7-94.9)

93.3 (91.7-94.9)

91.9 (89.3-94.5)

BSPLSDA

68.5 (65.1-71.9)

66.2 (62.5-69.9)

73.0 (69.8-76.2)

91.9 (90.4-93.4)

91.8 (90.2-93.4)

89.5 (87.8-91.2)

CF

81.1 (79.6-82.6)

82.2 (81.7-82.7)

88.1 (87.6-88.6)

99.9 (99.7-100)

99.9 (99.7-100)

99.9 (99.7-100)

GMU

80.3 (78.4-82.2)

80.8 (79.3-82.3)

88.6 (87.1-90.1)

99.2 (98.6-99.8)

99.2 (98.6-99.8)

98.8 (97.7-99.9)

Mogonet

81.6 (79.6-83.6)

81.4 (79.7-83.1)

84.0 (80.6-87.4)

97.7 (95.7-99.7)

97.6 (95.5-99.7)

95.8 (91.8-99.8)

Dynamics

83.3 (82.8-83.8)

83.7 (83.5-83.9)

88.5 (88.3-88.7)

99.9 (99.8-100)

99.9 (99.8-100)

99.9 (99.8-100)

MOCAT(Ours)

85.1 \(^*\) (84.4-85.8)

85.1 \(^*\) (84.1-86.1)

88.5 (88.0-89.0)

99.9 (99.8-100)

99.9 (99.8-100)

99.8 (99.3-100)

  1. Means and 95% confidence intervals (95% CIs) are presented, and the best results are in bold.The 95% CI is calculated using the t-distribution, with degrees of freedom set at \(n-1\), where n is the number of experiments conducted.
  2. Compared to the suboptimal model, the superior model is denoted by \(^*\) to indicate a statistically significant improvement (\(P<0.05\)) when using the two-sample t-test