[BioC] Interaction term logFC

[Michael Love]

hi Neetha,

You should look up a statistics reference for interaction terms in
linear models, or ask for an explanation from a statistician at your
institution. The interaction concept for linear models is the same as
here, although here we work with additive log fold changes, which
represent multiplication of fold changes.

Suppose  we have a design ~ A + B+ A:B, and A=0/1, B=0/1. This means
we are modeling a log fold change due to variable A=1 over A=0, and a
log fold change due to variable B=1 over B=0. The interaction term is
an additional log fold change when A=1 and B=1 beyond the main effect
for A and the main effect for B. If the log fold change for the
interaction term is 0, then we know: the fold change when A=1 and B=1
is simply the two main effect fold changes multiplied (the log fold
changes added).

Mike


On Wed, Jul 16, 2014 at 1:10 PM, Neetha [guest] <guest at bioconductor.org> wrote:
> Hi,
>
>
> I am using DESeq2 for my RNAseq data analysis. My design has two strains and two conditions. I did my analysis with DESeq2 v 1.4.5 with the design
> dds1 <- DESeqDataSetFromMatrix(countData=counts, colData=design, design=~ Colony + Treatment + Colony:Treatment)
> dds1 <- DESeq(dds1, test="LRT", reduced=Colony + Treatment)
>>resultsNames (dds1)
> [1] "Intercept"  "Colony_1_vs_2" "Treatment_1_vs_2"    "Colony1.Treatment1"
>
> "Colony1.Treatment1" is the interaction term. Am I right? I don't understand what the log2fold change represents. I read the vignette (last updated May 2014) as well several discussions online, but I am not sure I understand the concept. For those genes with FDR corrected p values below 0.05, when go back and check the VST counts/raw reads I see trends. Could you please help me understand what the log2foldchange means?
> Thanks in advance.
> Neetha
>
>
>  -- output of sessionInfo():
>
>> sessionInfo()
> R version 3.1.1 (2014-07-10)
> Platform: i386-w64-mingw32/i386 (32-bit)
>
> locale:
> [1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252
> [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
> [5] LC_TIME=English_United States.1252
>
> attached base packages:
> [1] parallel  stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] DESeq2_1.4.5            RcppArmadillo_0.4.320.0 Rcpp_0.11.2
> [4] GenomicRanges_1.16.3    GenomeInfoDb_1.0.2      IRanges_1.22.9
> [7] BiocGenerics_0.10.0     BiocInstaller_1.14.2
>
> loaded via a namespace (and not attached):
>  [1] annotate_1.42.0      AnnotationDbi_1.26.0 Biobase_2.24.0       DBI_0.2-7
>  [5] genefilter_1.46.1    geneplotter_1.42.0   grid_3.1.1           lattice_0.20-29
>  [9] locfit_1.5-9.1       RColorBrewer_1.0-5   RSQLite_0.11.4       splines_3.1.1
> [13] stats4_3.1.1         survival_2.37-7      tools_3.1.1          XML_3.98-1.1
> [17] xtable_1.7-3         XVector_0.4.0
>
>
> --
> Sent via the guest posting facility at bioconductor.org.