More powerful significant testing for time course gene expression data using functional principal component analysis approaches

Table 3 Comparison of the proposed method, EDGE and Sohn’s method for the single group test by FPR (proportion of falsely rejected hypotheses over the total number of rejected hypotheses) and sensitivity (proportion of true positives correctly identified)

			Proposed^a	Proposed^b	EDGE	Sohn’s
Π₀=0.05	FDR = 0.01	FPR	0.0016	0.0127	0.0100	0.0091
		Sensitivity	0.7606	0.6864	0.4742	0.5160
	FDR = 0.05	FPR	0.0180	0.0541	0.0459	0.0514
		Sensitivity	0.8440	0.7858	0.6378	0.6616
Π₀=0.2	FDR = 0.01	FPR	0.0003	0.0084	0.0079	0.0078
		Sensitivity	0.7716	0.7851	0.6210	0.6302
	FDR = 0.05	FPR	0.0070	0.0409	0.0363	0.0408
		Sensitivity	0.8644	0.8667	0.7652	0.7641
Π₀=0.5	FDR = 0.01	FPR	0.0002	0.0053	0.0041	0.0051
		Sensitivity	0.7731	0.8329	0.7021	0.7031
	FDR = 0.05	FPR	0.0025	0.0251	0.0238	0.0244
		Sensitivity	0.8709	0.9049	0.8285	0.8261

Proposed ^b uses (7) for computing p-values and Proposed ^a, EDGE and Sohn’s use (8).

ISSN: 1471-2105