Question

edgeR time series analysis question

0

Entering edit mode

Guest User ★ 13k

@guest-user-4897

Last seen 9.6 years ago

Hi, I have RNAseq data in replicates for 6 time points (0h, 6h, 12h, 24h, 96h, and 10 days). I want to do time course analysis of this data using GLM. Experimenting, I used a model matrix with an intercept term and the section on the ANOVA-like test for any differences. My question: Is the list of genes output based on contrasts between all the time points or just between each with the intercept(0h)? I hope this makes sense. Thanks so much! Best, Audra -- output of sessionInfo(): > sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-apple-darwin10.8.0 (64-bit) locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8 attached base packages: [1] tcltk parallel stats graphics grDevices utils [7] datasets methods base other attached packages: [1] maSigPro_1.34.0 DynDoc_1.40.0 widgetTools_1.40.0 [4] MASS_7.3-29 Biobase_2.22.0 BiocGenerics_0.8.0 [7] edgeR_3.4.2 limma_3.18.9 loaded via a namespace (and not attached): [1] Mfuzz_2.20.0 tkWidgets_1.40.0 tools_3.0.2 -- Sent via the guest posting facility at bioconductor.org.

RNASeq RNASeq • 2.0k views

ADD COMMENT • link updated 10.1 years ago by Gordon Smyth 50k • written 10.1 years ago by Guest User ★ 13k

score 0 · Answer 1 · 2014-03-09

Dear Audra,

I am assuming that you are following Section 3.2.6 (An ANOVA like test for any differences) of the edgeR User's Guide, and that you have used

lrt <- glmLRT(fit, coef=2:6)

If so, then the test is for all possible differences between the 6 time points.

The fact that the coefficients are fitted with time 0 as the reference level makes no difference. The test would be identical if another test of 5 contrasts was used.

Best wishes
Gordon