[BioC] Testing for no change in RNA-seq data?

[Ryan C. Thompson]

Are there equivalence tests that do not require the assumption of 
normality? Or should I transform with limma::voom and pretend it's normal?

-Ryan

On 03/20/2013 05:15 PM, Albyn Jones wrote:
> see "tests of equivalence", a common one being the TOST,
> aka two one-sided tests.  you need to have a substantive
> definition of equivalence!
>
> albyn
>
> On 2013-03-20 17:05, Ryan C. Thompson wrote:
>> Hi all,
>>
>> So, we have several great packages for testing for evidence of
>> differential expression using RNA-seq counts, and I am happily using
>> those. However, I would like to know if there is any method for
>> testing for evidence of equivalent expression. That is, I would like
>> to show with some level of confidence that treatment X has no (or
>> negligible) effect on gene Y. Is there any way to conduct such a test?
>>
>> I know I can test for differential expression and see if the gene
>> list comes up empty, but lack of evidence for differential expression
>> is not the same as evidence for equivalent expression. So, are there
>> any packages for doing equivalence testing on gene expression data? If
>> it exists for limma, then I'm happy to run my data through voom to
>> analyze it if necessary.
>>
>> Thanks,
>>
>> -Ryan Thompson
>>
>> _______________________________________________
>> 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
>
> _______________________________________________
> 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