Hi, this is my first post on BiocR, I learned a lot by reading existing posts. I am trying to use DESeq to infer differential gene expression in a 2-factor design of an RNA-seq expt. Factor 1 (strain) and Factor 2 (condition) each have 4 levels; each level has 3 bioreps totaling to 48 RNA-seq samples. I got good results from within-factor analysis, and am now trying to identify an interaction effect of factor1 and factor2 on gene expression. This is the code i used: > fit1 = fitNbinomGLMs( cdsFull, count ~ strain + condition + strain : condition ) > fit0 = fitNbinomGLMs( cdsFull, count ~ strain + condition ) > pvalsGLM = nbinomGLMTest( fit1, fit0 ) > padjGLM = p.adjust( pvalsGLM, method='BH' ) > DEresults <- transform( fit1, pval=pvalsGLM, padj=padjGLM ) > head( DEresults[ order( DEresults$padj ), ] ) Attached is the output of head. What I am asking help with is how to determine which of these interaction comparisons are significant. Do I need to compare each factor1 level with each factor2 level, ie pairwise strain : condition comparisons, or is there a way to extract this from the existing data? Can this be done in DESeq or should I be looking at another package? Your insight is greatly appreciated. thanks galeb -------------- next part -------------- X.Intercept. strain2 strain3 strain4 condition2 condition3 condition4 strain2.condition2 strain3.condition2 strain4.condition2 strain2.condition3 strain3.condition3 strain4.condition3 strain2.condition4 strain3.condition4 strain4.condition4 deviance converged pval padj gene_0477 9.3980690498 -1.6791314581 -7.002798797 -1.7106317853 0.1292043759 1.004851704 0.9593594984 0.4460683887 1.3333587254 0.5430856124 1.0467756694 4.3566446059 1.4022027955 1.0247376795 4.0497624618 1.5042551291 66.9551796174 TRUE 0 0 gene_0899 8.8283772691 1.3201257541 3.6966878444 2.0598916209 -0.2863640307 -3.1862934861 -3.6320668852 -0.433159713 0.1183208245 -0.3313504214 -1.3089286975 1.1287940418 -1.8929916603 -0.6393882394 1.6738458015 -1.0571110161 36.5611791438 TRUE 0 0 gene_3193 9.504666246 -0.1170027097 -4.1303583211 -0.6022719781 0.09498906 0.526477533 0.574614423 0.0826986394 0.1735329192 0.2855168364 0.2909434165 2.5168302886 0.6268837601 0.1252186408 2.4662137035 0.5265930419 20.2189071609 TRUE 0 0 gene_3190 9.7134519192 -1.2708138946 -4.6197170035 -1.4720599481 -0.2505335899 0.3404216725 0.5225965378 0.4759426442 0.0465311861 0.536341491 0.833730625 2.3164668939 0.8533719617 0.5648146485 2.0587035777 1.0514146972 32.5110294464 TRUE 2.19E-014 2.22E-011 gene_3192 8.3830050576 -0.6638001747 -4.1436143915 -0.952456834 0.0808601394 0.5629341824 0.4775418862 0.2657938573 -0.1631299939 0.3539415549 0.4834241261 2.2290320256 0.7099500394 0.4530668099 2.3089386737 0.7734138303 29.3662998474 TRUE 8.77E-014 7.13E-011 gene_0031 9.3983124954 0.4919132644 -0.1754119696 0.5397965786 0.3580852136 -2.7711159019 -3.7408103493 0.1876451807 -0.0149511116 -0.1168719638 -0.6959492519 1.3440430005 -0.6036571287 -0.159308005 1.8053857034 -0.4823270854 49.8376020899 TRUE 1.22E-013 8.30E-011

Hi, just to follow up with my first post, dealing with interaction term analysis of RNA-seq 2-factor expt. In case anyone is interested, package egdeR offers explicit examples how to evaluate potential interaction effects in multi-factorial design. Section 3.3 specifically. cheers galeb Galeb S. Abu-Ali, DVM, PhD ORISE Fellow Division of Molecular Biology Center for Food Safety and Applied Nutrition U.S. Food and Drug Administration 8301 Muirkirk Road Laurel, MD 20708 (240) 402 3695 galeb.abu-ali@fda.hhs.gov<https: webmail.fda.gov="" owa="" redir.aspx?c="gd8_swQGmEyTnnsl5KrApdNupyzkR9AIlS4rHtgTQAkWp01Gbec94M2etJR-0PyoCUJ-" ek2qiqm.&url="mailto%3agaleb.abu-ali%40fda.hhs.gov"> > sessionInfo() R version 3.0.1 (2013-05-16) Platform: x86_64-pc-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 [7] LC_PAPER=C LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] edgeR_3.2.3 limma_3.16.5 loaded via a namespace (and not attached): [1] tools_3.0.1 [[alternative HTML version deleted]]