Skip to main content

Table 1 Results for GWAS data simulated from LMM with \(n = 400, p = 225{,}000, \kappa = 1\), and \(\sigma ^2 = 0.2\)

From: BGWAS: Bayesian variable selection in linear mixed models with nonlocal priors for genome-wide association studies

Nominal FDR

Method

\(\beta ^{(1)} = 0.1\)

\(\beta ^{(1)} = 0.4\)

\(\beta ^{(1)} = 1.6\)

TP

FP

F1

Time (s)

TP

FP

F1

Time (s)

TP

FP

F1

Time (s)

0.05

SMA-Approx.

5.2

6.9

0.38

4

4.6

4.1

0.39

4

3.9

36.4

0.14

4

SMA-Exact

5.2

7.0

0.38

103

4.6

4.2

0.39

104

3.9

36.9

0.14

93

NP, \(\tau = 0.348\)

4.2

0.6

0.42

37

3.1

0.1

0.34

17

4.0

0.0

0.42

29

NP, \(\tau = 0.022\)

6.1

0.8

0.55

35

6.3

0.6

0.57

30

4.1

0.0

0.43

35

NP, \(\tau\) estimated

6.4

0.8

0.57

40

6.7

0.8

0.60

36

4.1

0.0

0.43

37

0.1

SMA-Approx.

5.5

8.1

0.39

4

5.3

5.2

0.42

4

4.0

41.4

0.13

4

SMA-Exact

5.6

8.2

0.39

103

5.3

5.3

0.42

104

4.0

41.9

0.13

93

NP, \(\tau = 0.348\)

4.6

0.7

0.45

32

3.9

0.2

0.40

33

4.0

0.0

0.42

31

NP, \(\tau = 0.022\)

6.4

0.9

0.57

39

6.8

0.8

0.60

35

4.4

0.0

0.45

37

NP, \(\tau\) estimated

6.6

0.9

0.59

48

7.0

1.1

0.61

45

4.4

0.0

0.45

40

  1. In this table, there are 15 causal SNPs. The regression coefficients of the 15 causal SNPs are \(\varvec{\beta }= (\beta ^{(1)}, 0.4, 0.4, 0.4, \beta ^{(1)}, 0.4, 0.4, 0.4, \beta ^{(1)}, 0.4, 0.4, 0.4, \beta ^{(1)}, 0.4, 0.4)^\top\). TP indicates Average number of True Positives, FP is Average number of False Positives, and F1 is the Average F1 score. Average Performance of each method over 50 datasets for each setting