Skip to main content

Table 4 Simulation study 1, scenario 1 with nX=nY=30 and 100 runs

From: Identification of differentially expressed genes by means of outlier detection

Δ

DE

10f0

C i

\(\bar {n}_{i}\)

\(\overline {OR}\)

\(\overline {FP}\)

\(\overline {dFP}\)

\(\overline {FPC}\)

    

(sd)

(sd)

(sd)

(sd)

(sd)

   

\(C_{{1}^{*}}\)

30.7

45.7

0.5

0.4

0.2

    

(7.9)

(6.7)

(0.3)

(0.3)

(0.9)

 

50

8.5

\(C_{{2}^{*}}\)

21.8

19.4

5.4

7.5

20.7

    

(5.9)

(5.0)

(2.1)

(4.4)

(20.9)

   

C 3

32.9

8.4

9.2

29.0

93.7

    

(6.3)

(0.6)

(0.2)

(3.0)

(5.8)

   

\(C_{1}^{*}\)

34.1

46.5

0.0

0.0

0.0

    

(11.3)

(6.3)

(0.1)

(0.1)

(0.2)

1.5

100

8.5

\(C_{2}^{*}\)

56.2

22.5

1.5

1.8

1.6

    

(10.3)

(2.7)

(0.67)

(9.0)

(2.0)

   

C 3

32.1

9.1

8.7

21.5

68.9

    

(5.6)

(0.5)

(0.35)

(2.0)

(8.9)

   

\(C_{1}^{*}\)

59.3

29.5

0.0

0.0

0.0

    

(12.5)

(2.8)

(0.0)

(0.0)

(0)

 

200

8.5

\(C_{2}^{*}\)

113.0

14.8

0.5

0.8

0.4

    

(13.0)

(1.0)

(0.2)

(0.2)

(0.6)

   

C 3

25.3

8.0

6.0

9.9

14.9

    

(6.5)

(0.5)

(0.9)

(1.8)

(8.2)

   

\(C_{1}^{*}\)

22.3

86.3

0.0

0.0

0.0

    

(9.2)

(17.1)

(0.1)

(0.0)

(0)

 

50

8.0

\(C_{2}^{*}\)

27.8

42.0

1.4

1.5

5.8

    

(6.9)

(9.2)

(2.0)

(3.3)

(18.0)

   

C 3

36.0

9.1

9.2

27.8

97.0

    

(5.3)

(0.61)

(0.19)

(2.0)

(2.7)

   

\(C_{1}^{*}\)

35.5

74.7

0.0

0.0

0.0

    

(7.9)

(8.1)

(0.0)

(0.0)

(0)

2

100

8.0

\(C_{2}^{*}\)

63.1

37.3

0.4

0.4

0.3

    

(7.9)

(3.1)

(0.2)

(0.2)

(0.7)

   

C 3

26.0

8.8

9.3

24.9

93.7

    

(4.7)

(0.6)

(0.3)

(1.9)

(4.5)

   

\(C_{1}^{*}\)

69.2

46.6

0.0

0.0

0.0

    

(14.1)

(3.9)

(0.0)

(0.0)

(0)

 

200

8.0

\(C_{2}^{*}\)

122.9

23.8

0.2

0.2

0.0

    

(17.0)

(2.4)

(0.1)

(0.1)

(0.3)

   

C 3

13.1

8.9

8.1

13.0

58.0

    

(17.3)

(2.4)

(2.0)

(4.1)

(23.1)

   

\(C_{1}^{*}\)

18.9

191.2

0.0

0.0

0.0

    

(6.6)

(22.7)

(0.0)

(0.0)

(0)

 

50

7.2

\(C_{2}^{*}\)

31.5

99.9

0.29

0.4

2.1

    

(4.9)

(16.1)

(1.2)

(2.6)

(14.1)

   

C 3

37.0

9.0

9.2

27.7

99.8

    

(5.1)

(0.6)

(0.19)

(1.5)

(0.8)

   

\(C_{1}^{*}\)

38.6

155.7

0.0

0.0

0.0

    

(10.0)

(15.1)

(0.0)

(0.0)

(0)

3

100

7.1

\(C_{2}^{*}\)

61.3

83.4

0.0

0.0

0.0

    

(10.0)

(6.6)

(0.1)

(0.0)

(0)

   

C 3

25.1

8.7

9.3

25.8

99.6

    

(4.5)

(0.49)

(0.26)

(1.7)

(1.1)

   

\(C_{1}^{*}\)

74.1

95.8

0

0

0.0

    

(16.4)

(8.2)

(0)

(0)

(0)

 

200

7.1

\(C_{2}^{*}\)

115.5

53.1

0.0

0.0

0.0

    

(21.3)

(6.6)

(0.0)

(0.0)

(0)

   

C 3

16.1

12.5

8.3

16.4

86.0

    

(24.8)

(11.9)

(3.2)

(6.8)

(33.6)

  1. Evaluation of the second step of the ORdensity method with α=0.05. In the first two columns, delta (Δ) values and number of total simulated DE genes. In column 3, the 10×f0 values where f0 is the average proportion of permuted cases in sets Ui. In column 4, the “*” indicates the clusters considered by the procedure. Columns 5–8 contain for each cluster: the mean number of genes (\(\bar {n}_{i}\)), the mean of OR values (\(\overline {OR}\)), the mean of FP values (\(\overline {FP}\)), the mean of dFP values (\(\overline {dFP}\)). In the last column the mean of False Positives genes per cluster in % (\(\overline {FPC}\)). Corresponding standard deviations are in brackets