Skip to main content

Table 4 Performance evaluation under Study II (Setting II), (G=5000, 500 DE gene) #

From: Confident difference criterion: a new Bayesian differentially expressed gene selection algorithm with applications

Method

CDE

CCDE

CCEE

FNR

FPR

FDR

FNDR

twocri.(γ 0=0.4)

654.6(5.4)

460.9(2.3)

4306.3(5.2)

0.078(0.005)

0.043(0.001)

0.295(0.006)

0.009(0.000)

(γ 0=0.6)

490.0(3.9)

434.1(2.4)

4444.2(3.4)

0.132(0.005)

0.012(0.001)

0.114(0.006)

0.014(0.001)

(ζ=0.05)

415.7(5.0)

400.6(3.5)

4484.9(2.7)

0.199(0.007)

0.003(0.001)

0.036(0.006)

0.022(0.001)

edgeR 1(ζ=0.05)

420.0(8.6)

411.7(8.4)

4491.7(2.8)

0.177(0.017)

0.002(0.001)

0.020(0.001)

0.020(0.002)

edgeR 2(ζ=0.05)

399.4(10.2)

386.3(9.8)

4486.9(4.6)

0.227(0.020)

0.003(0.001)

0.033(0.011)

0.025(0.002)

DESeq(ζ=0.05)

443.6(15.9)

409.3(15.1)

4465.8(5.3)

0.181(0.030)

0.008(0.001)

0.077(0.011)

0.020(0.003)

BaySeq(ζ=0.05)

331.0(15.5)

327.0(14.9)

4496.0(2.2)

0.346(0.030)

0.001(0.000)

0.012(0.006)

0.037(0.003)

NBPSeq(ζ=0.05)

422.3(7.9)

412.7(7.9)

4490.4(3.1)

0.175(0.016)

0.002(0.001)

0.023(0.007)

0.019(0.002)

EBSeq(ζ=0.05)

332.9(14.0)

248.4(11.1)

4415.6(9.4)

0.503(0.022)

0.019(0.002)

0.253(0.023)

0.054(0.002)

NOISeq(ζ=0.05)

196.8(11.0)

191.4(10.8)

4494.6(2.5)

0.617(0.022)

0.001(0.001)

0.028(0.013)

0.064(0.002)

SAMSeq(ζ=0.05)

274.3(15.7)

212.1(7.7)

4437.8(10.9)

0.576(0.015)

0.014(0.002)

0.226(0.029)

0.061(0.001)

TSPM(ζ=0.05)

129.9(10.5)

80.1 (8.9)

4450.2(7.1)

0.840(0.018)

0.011(0.002)

0.383(0.046)

0.086(0.002)

  1. #Empirical estimates of the standard deviation were reported in the parentheses
  2. edgeR 1 estimates common dispersion parameter for all tags; edgeR 2 estimats tag-wise dispersion parameters
  3. ζdenotes the false discovery rate