[BioC] DESeq2: detect significantly increasing across an ordinal variable

[Michael Love]

hi Dan,

The simplest way to define this set of genes would be to contrast
medium vs low, and then high vs medium. Then take the intersection of
those with both positive log fold change and low FDR in both results
tables. You can perform these contrasts either after the LRT or after
a default DESeq() call.

Mike

On Thu, Jul 10, 2014 at 6:16 AM, Daniel Brewer (BIO) <D.Brewer at uea.ac.uk> wrote:
> I have an RNASeq dataset that has samples from three conditions: low, medium and high.  What I would like to find is genes that have expression that either increases or decreases with the condition.  I have been using DESeq2 to do differential analysis and so was wondering if there is a way to do this.  I tried to use the log-ratio-test as indicated in the time-series section of the vignette, but that is a  "test for any di erences over multiple time points”.
>
> Any ideas? Thanks
>
> Dan
>
> ----
> Dr Daniel Brewer
> Senior Bioinformatics Officer | Cancer Genetics
> School of Biological Sciences | University of East Anglia | Norwich Research Park | Norwich | NR4 7TJ
>
> Tel: +44 (0) 1603 592906 | Email: d.brewer at uea.ac.uk
>
> _______________________________________________
> 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