[BioC] RankProd related question

Sean Davis sdavis2 at mail.nih.gov
Thu Oct 6 12:37:49 CEST 2011


Hi, Wendy.

The problem here is not with RankProd or any other statistical method.
 Unfortunately, when you have samples that are perfectly segregated
onto two different platforms, it is not possible to perform a
meaningful statistical analysis to test the null hypothesis.  The
biological signal (difference between two cell types) is perfectly
confounded with the difference between technologies.  I believe the
error message suggests that a remedy would be to have at least some of
each group on each platform; if you do not have the option to do that
(again, applies for any statistical testing procedure), you will
probably need to rethink your experimental design.

Hope that helps.

Sean


On Wed, Oct 5, 2011 at 11:00 PM, Wendy Qiao <wendy2.qiao at gmail.com> wrote:
> Hi all,
>
> I need to analyze a combined dataset of microarrays from two sources. The
> arrays for the first cell type is from source1 and the arrays for the other
> cell types are from source2. I would like to use the advanced rank product
> method to identify the unregulated genes. My "origin" vector and sample
> vector are as following. However, by using the following code, I got an
> error message as highlighted. I think that the arrays for cell type 1 must
> from two sources. Does it mean that I cannot use the advanced RankProd? Does
> anyone know the major difference between the normal and advanced RankProd
> methods?
>
> Thank you in advance,
> Wendy
>
>
> mergeD.origin   # origin vector
>  [1] 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
> 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
> [54] 2 2 2 2 2 2 2 2
>
> mergeD.cl   # sample vector
>  [1] 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> [54] 0 0 0 0 0 0 0 0
>
> RP.adv.out <- RPadvance(mergeD,  mergeD.cl, mergeD.origin, num.perm = 100,
> logged = FALSE, gene.names = rownames(mergeD), rand = 123)
>  The data is from  2 different origins
>
> Rank Product analysis for two-class case
>
> Error in OriginxyCall(data, cl, origin) :
>  Error: data from different origins should contain data from both classes
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> 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
>



More information about the Bioconductor mailing list