[BioC] Interaction term logFC

[James W. MacDonald]

Hi Neetha,

The interaction term is (algebraically) something like

(colony1_treatment_1 - colony_1_treatment_2) - (colony_2_treatment_1 - 
colony_2_treatment_2)

In other words, you are trying to see if the different treatments affect 
the two colonies differently. And because of the way it is constructed, 
the interaction term fold change isn't by itself that easily interpreted.

As an example, if I just substitute fake numbers into the above formula, 
all of these result in a logFC of 2:

(3 - 3) - (1 - 3)
(3 - 1) - (3 - 3)
(3 - 2) - (2 - 3)

There are many other ways a logFC of 2 can arise, so you really want to 
look at a plot of the logCPM values to see the underlying pattern. This 
is where the ReportingTools package comes in. You can easily make an 
HTML table that has little plots in each row that show the 
directionality of expression for all four groups. See the vignette for 
more information:

http://bioconductor.org/packages/release/bioc/vignettes/ReportingTools/inst/doc/rnaseqAnalysis.pdf

Best,

Jim



On 7/16/2014 1:10 PM, Neetha [guest] 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.
>
> _______________________________________________
> 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
>

-- 
James W. MacDonald, M.S.
Biostatistician
University of Washington
Environmental and Occupational Health Sciences
4225 Roosevelt Way NE, # 100
Seattle WA 98105-6099