Table 1

The proportion of DE genes (p1 = 0.05), AUC, and pAUC with a FPR ≤ 0.2 estimated by the t-test, Limma, SAM, and EB approaches (EB-LNN, eGG, eLNN, GaGa) and the β -EB approach averaged over 50 simulated datasets: the case of large sample
t Limma SAM eGG eLNN GaGa EB-LNN β -EB
In absence of outliers
p1 - - - 0.0488 0.0458 0.0494 0.0496 0.0482
- - - (0.0010) (0.0009) (0.0010) (0.0010) (0.0013)
AUC 0.9861 0.9861 0.9862 0.9848 0.9734 0.9879 0.9892 0.9890
(0.0020) (0.0021) (0.0020) (0.0019) (0.0030) (0.0017) (0.0015) (0.0016)
pAUC 0.1929 0.1934 0.1924 0.1925 0.1894 0.1940 0.1941 0.1940
(0.0008) (0.0008) (0.0008) (0.0008) (0.0011) (0.0006) (0.0007) (0.0007)
In presence of 10% contaminated genes with mild outliers
p1 - - - 0.0807 0.1053 0.1008 0.0649 0.0504
- - - (0.0013) (0.0012) (0.0014) (0.0013) (0.0014)
AUC 0.9649 0.9661 0.9699 0.9515 0.9396 0.9524 0.9621 0.9870
(0.0031) (0.0030) (0.0029) (0.0030) (0.0026) (0.0052) (0.0020) (0.0019)
pAUC 0.1826 0.1830 0.1844 0.1696 0.1577 0.1649 0.1724 0.1924
(0.0012) (0.0012) (0.0012) (0.0012) (0.0009) (0.0008) (0.0009) (0.0008)
In presence of 10% contaminated genes with extreme outliers
p1 - - - 0.0834 0.1076 0.1043 0.0599 0.0489
- - - (0.0015) (0.0012) (0.0014) (0.0013) (0.0014)
AUC 0.9692 0.9695 0.9676 0.9488 0.9333 0.9422 0.9601 0.9880
(0.0031) (0.0031) (0.0028) (0.0034) (0.0030) (0.0064) (0.0019) (0.0017)
pAUC 0.1842 0.1844 0.1834 0.1684 0.1542 0.1610 0.1617 0.1931
(0.0012) (0.0012) (0.0011) (0.0010) (0.0010) (0.0009) (0.0010) (0.0007)
In presence of 20% contaminated genes with mild outliers
p1 - - - 0.1275 0.1693 0.1565 0.0946 0.0521
- - - (0.0016) (0.0014) (0.0016) (0.0018) (0.0016)
AUC 0.9405 0.9415 0.9430 0.9147 0.8984 0.9085 0.9502 0.9850
(0.0041) (0.0041) (0.0030) (0.0028) (0.0025) (0.0026) (0.0021) (0.0017)
pAUC 0.1728 0.1727 0.1723 0.1409 0.1214 0.1320 0.1601 0.1904
(0.0014) (0.0014) (0.0011) (0.0009) (0.0007) (0.0006) (0.0014) (0.0007)
In presence of 20% contaminated genes with extreme outliers
p1 - - - 0.1260 0.1735 0.1614 0.0869 0.0502
- - - (0.0023) (0.0014) (0.0015) (0.0015) (0.0014)
AUC 0.9465 0.9460 0.9455 0.9112 0.8910 0.8980 0.9421 0.9869
(0.0040) (0.0040) (0.0034) (0.0035) (0.0034) (0.0035) (0.0028) (0.0017)
pAUC 0.1733 0.1721 0.1720 0.1391 0.117 0.1282 0.1539 0.1923
(0.0014) (0.0014) (0.0012) (0.0012) (0.0010) (0.0009) (0.0016) (0.0008)

The numbers in parentheses are the standard errors for the 50 simulation trails.

Hossain Mollah et al.

Hossain Mollah et al. BMC Bioinformatics 2012 13:135   doi:10.1186/1471-2105-13-135

Open Data