Table 3

Percentages of significant cases detected (Detected) and false discovery rate (FDR) after multitest adjustment when the p-values come from families of homogeneity tests where some (% effect) of the alternative hypotheses were true.

Weak

SB

BH

SGoF




N

% effect

S

Significant

Detected

FDR

Detected

FDR

Detected

FDR


20

5%

100

5.7

0.10

73

0.10

72

0.10

77

20

5%

1000

5.7

0.01

53

0.01

57

0.07

78

20

5%

10000

5.7

0.00

65

0.00

61

0.25

55

20

10%

100

6.2

0.07

51

0.08

51

0.18

68

20

10%

1000

6.1

0.01

41

0.01

40

0.19

56

20

10%

10000

6.1

0.00

40

0.00

33

0.68

33

20

20%

100

6.8

0.14

39

0.16

38

0.30

52

20

20%

1000

7.0

0.01

24

0.01

25

0.70

35

20

20%

10000

7.0

0.00

16

0.00

16

1.58

19


Mean

0.04

44.7

0.04

43.7

0.45

52.6

SD

0.052

18.17

0.058

18.06

0.482

20.37


40

5%

100

6.0

0.14

45

0.15

46

0.13

69

40

5%

1000

6.0

0.02

31

0.02

33

0.15

57

40

5%

10000

6.0

0.00

22

0.01

23

0.61

37

40

10%

100

7.0

0.21

26

0.25

28

0.33

51

40

10%

1000

7.0

0.03

18

0.04

20

0.71

38

40

10%

10000

7.0

0.01

12

0.01

14

1.56

17

40

20%

100

8.9

0.34

20

0.45

21

1.02

32

40

20%

1000

8.9

0.05

8

0.10

9

2.50

21

40

20%

10000

8.9

0.01

5

0.04

7

3.44

6


Mean

0.09

20.8

0.12

22.3

1.16

36.4

SD

0.117

12.34

0.148

12.23

1.143

20.19


Weak: The alternative hypothesis is weak. N: sample size. % effect: % of real true alternatives. S: number of tests. Significant: % of significants before adjustment. Detected: % of significant tests after adjustment. FDR: false discovery rate. SB: Sequential Bonferroni. BH: Benjamini and Hochberg. SGoF: Sequential Goodness of Fit.

Carvajal-Rodríguez et al. BMC Bioinformatics 2009 10:209   doi:10.1186/1471-2105-10-209

Open Data