DESeq: Hypothesis testing in multifactor design
1
0
Entering edit mode
Guest User ★ 13k
@guest-user-4897
Last seen 10.5 years ago
Dear Community, I have a question about the hypothesis testing of the two-way interaction terms in a multifactor design which includes three factors: A, B and C. When I tested the three-way interaction I used the full and reduced models as below for nbinomGLMTest(): Full: count ~ A+B+C+A:B+A:C+B:C+A:B:C Reduced: count ~ A+B+C+A:B+A:C+B:C Now comes my question, when I want to test the effect of two-way interaction terms, i.e., A:B, A:C or B:C, what should be my full and reduced models? For example, when I want to the test the effect of A:B, what should be my full and reduced models for nbinomGLMTest() using DESeq pacakge? Best, Yanzhu -- output of sessionInfo(): sessionInfo() R version 3.1.0 (2014-04-10) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 [4] LC_NUMERIC=C LC_TIME=English_United States.1252 attached base packages: [1] parallel stats graphics grDevices utils datasets methods base other attached packages: [1] DESeq_1.16.0 lattice_0.20-29 locfit_1.5-9.1 Biobase_2.24.0 BiocGenerics_0.10.0 edgeR_3.6.1 limma_3.20.1 loaded via a namespace (and not attached): [1] annotate_1.42.0 AnnotationDbi_1.26.0 DBI_0.2-7 genefilter_1.46.0 geneplotter_1.42.0 GenomeInfoDb_1.0.2 [7] grid_3.1.0 IRanges_1.22.6 MASS_7.3-31 RColorBrewer_1.0-5 RSQLite_0.11.4 splines_3.1.0 [13] stats4_3.1.0 survival_2.37-7 tools_3.1.0 XML_3.98-1.1 xtable_1.7-3 -- Sent via the guest posting facility at bioconductor.org.
DESeq DESeq • 1.1k views
ADD COMMENT
0
Entering edit mode
@mikelove
Last seen 6 hours ago
United States
hi Yanzhu, Note that we recommend users switch to using DESeq2, which also has the likelihood ratio test you are using, and is faster and more sensitive. The pipeline would look like: DESeq(dds, test="LRT", reduced=~ A+B+C+A:B+A:C+B:C) for your first example. For your question, the terms of the reduced model should be contained within the full model. Still there are a number of models which satisfy this requirement, e.g. for testing B:C, you could use A+B+C+A:B+A:C+B:C and A+B+C+A:B+A:C as full and reduced respectively. Or you could use A+B+C+B:C and A+B+C. The importance of these other interaction terms depends on context, whether they are very explanatory or not. Mike On Tue, Jun 10, 2014 at 11:21 AM, yanzhu [guest] <guest at="" bioconductor.org=""> wrote: > Dear Community, > > I have a question about the hypothesis testing of the two-way interaction terms in a multifactor design which includes three factors: A, B and C. > > When I tested the three-way interaction I used the full and reduced models as below for nbinomGLMTest(): > Full: count ~ A+B+C+A:B+A:C+B:C+A:B:C > Reduced: count ~ A+B+C+A:B+A:C+B:C > > Now comes my question, when I want to test the effect of two-way interaction terms, i.e., A:B, A:C or B:C, what should be my full and reduced models? For example, when I want to the test the effect of A:B, what should be my full and reduced models for nbinomGLMTest() using DESeq pacakge? > > > Best, > > > > Yanzhu > > > -- output of sessionInfo(): > > sessionInfo() > R version 3.1.0 (2014-04-10) > Platform: x86_64-w64-mingw32/x64 (64-bit) > > locale: > [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 > [4] LC_NUMERIC=C LC_TIME=English_United States.1252 > > attached base packages: > [1] parallel stats graphics grDevices utils datasets methods base > > other attached packages: > [1] DESeq_1.16.0 lattice_0.20-29 locfit_1.5-9.1 Biobase_2.24.0 BiocGenerics_0.10.0 edgeR_3.6.1 limma_3.20.1 > > loaded via a namespace (and not attached): > [1] annotate_1.42.0 AnnotationDbi_1.26.0 DBI_0.2-7 genefilter_1.46.0 geneplotter_1.42.0 GenomeInfoDb_1.0.2 > [7] grid_3.1.0 IRanges_1.22.6 MASS_7.3-31 RColorBrewer_1.0-5 RSQLite_0.11.4 splines_3.1.0 > [13] stats4_3.1.0 survival_2.37-7 tools_3.1.0 XML_3.98-1.1 xtable_1.7-3 > > > -- > Sent via the guest posting facility at bioconductor.org. > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
ADD COMMENT

Login before adding your answer.

Traffic: 528 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6