Hi everyone, Does nbinomialTest in DESeq calculate a test statistic? According to DESeq documentation, "nbinomTest calculates a p value by summing up the probabilities of all per-group count sums a and b that sum up to the observed count sum kiS and are more extreme than the observed count sums kiA and kiB. " The journal we are submitting our work to requires that we state both the test statistic and the p-value in our results. Since our analysis was performed using DESeq, I need to know how to extract the test statistic or make a statement about how the p-value is calculated. Any help/advice on this matter would be greatly appreciated. Thanks Sarven -- output of sessionInfo(): > sessionInfo() R version 3.1.0 Patched (2014-04-24 r65479) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 [7] LC_PAPER=en_US.UTF-8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] parallel stats graphics grDevices datasets utils methods [8] base other attached packages: [1] DESeq_1.16.0 lattice_0.20-29 locfit_1.5-9.1 [4] Biobase_2.24.0 BiocGenerics_0.10.0 loaded via a namespace (and not attached): [1] annotate_1.42.0 AnnotationDbi_1.26.0 DBI_0.2-7 [4] genefilter_1.46.1 geneplotter_1.42.0 GenomeInfoDb_1.0.2 [7] grid_3.1.0 IRanges_1.22.6 RColorBrewer_1.0-5 [10] RSQLite_0.11.4 splines_3.1.0 stats4_3.1.0 [13] survival_2.37-7 tools_3.1.0 XML_3.98-1.1 [16] xtable_1.7-3 -- Sent via the guest posting facility at bioconductor.org.

Dear Sarven depending on how one wants to view it, the test statistic is either: - (a) bivariate and given by the pair of observed count sums (kiA, kiB), - (b) that p-value, itself. The spirit of the journal guidelines would probably best satisfied by stating either (a) or the ratio kiA/kiB. Best wishes Wolfgang Il giorno 08 Sep 2014, alle ore 18:15, Sarven Sabunciyan [guest] <guest at="" bioconductor.org=""> ha scritto: > Hi everyone, > > Does nbinomialTest in DESeq calculate a test statistic? According to DESeq documentation, "nbinomTest calculates a p value by summing up the probabilities of all per-group count sums a and b that sum up to the observed count sum kiS and are more extreme than the observed count sums kiA and kiB. " > > The journal we are submitting our work to requires that we state both the test statistic and the p-value in our results. Since our analysis was performed using DESeq, I need to know how to extract the test statistic or make a statement about how the p-value is calculated. Any help/advice on this matter would be greatly appreciated. > > Thanks > > Sarven > > -- output of sessionInfo(): > >> sessionInfo() > R version 3.1.0 Patched (2014-04-24 r65479) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C > [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 > [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 > [7] LC_PAPER=en_US.UTF-8 LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C > > attached base packages: > [1] parallel stats graphics grDevices datasets utils methods > [8] base > > other attached packages: > [1] DESeq_1.16.0 lattice_0.20-29 locfit_1.5-9.1 > [4] Biobase_2.24.0 BiocGenerics_0.10.0 > > loaded via a namespace (and not attached): > [1] annotate_1.42.0 AnnotationDbi_1.26.0 DBI_0.2-7 > [4] genefilter_1.46.1 geneplotter_1.42.0 GenomeInfoDb_1.0.2 > [7] grid_3.1.0 IRanges_1.22.6 RColorBrewer_1.0-5 > [10] RSQLite_0.11.4 splines_3.1.0 stats4_3.1.0 > [13] survival_2.37-7 tools_3.1.0 XML_3.98-1.1 > [16] 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