BMC Genomics

Table 2 Estimated false discovery rate (FDR) of null hypotheses using different combinations of the three groups

From: Correspondence regarding "Effect of active smoking on the human bronchial epithelium transcriptome"

	null dataset			actual dataset
null hypothesis	≥20 TPM	p ≤ 0.05	fold-change ≥ 2	≥20 TPM	p ≤ 0.05	fold-change ≥ 2	FDR
N = C	7406	418	195	7764	885	609	47.2
N = F	7323	384	157	7547	765	447	50.2
F = C	7102	416	92	7318	895	433	46.5
(N, F) = C	7726	411	82	8148	959	460	42.8
N = (F, C)	7726	382	67	8148	836	475	45.7
(N, C) = F	7726	388	157	8148	818	314	47.4

The estimated false discovery rate (FDR) of different null hypotheses when grouping the 24 SAGE libraries of Chari et al. into never, former, and current smokers. A p-value cutoff of 0.05 was used, as in the original paper.

Back to article page

ISSN: 1471-2164

Contact us

Submission enquiries: bmcgenomics@biomedcentral.com
General enquiries: ORSupport@springernature.com