Skip to main content

Table 2 Concordance index and log-rank test of all models

From: Integration and comparison of different genomic data for outcome prediction in cancer

Cancer type

Algorithm

EXPR

MIRNA

CNA

MUT

MERGE

OV

CPSO

66b

61b

64c

10c

65c

NFS

60a

53

56b

11c

63c

LASSO

68c

62c

64c

-

68c

Average

65

59

61

10

65

LUAD

CPSO

74b

70

74b

52c

75b

NFS

71b

73b

65a

29b

64

LASSO

72c

75c

66c

52c

78c

Average

72

72

68

44

72

BRCA

CPSO

85c

82c

92

38c

83c

NFS

79

76

70

28c

84

LASSO

81c

80b

83c

53c

86c

Average

82

80

82

40

84

GBM

CPSO

63c

59c

57b

16c

59

NFS

60c

61c

58b

3b

63c

LASSO

60c

61c

53c

5

61c

Average

61

61

56

8

61

Overall

CPSO

72

68

72

29

71

NFS

67

66

62

18

69

LASSO

70

70

66

37

73

Average

70

68

67

27

71

  1. a,b,cIndicate models whose Kaplan-Meier curves were statistically different at 0.05, 0.01, and 0.001 level respectively using the log-rank test. For this, the population was split by the median using the prognostic index (linear predictor of the Cox model). “-” indicates that no models were generated