[BioC] collapseRows function

Alex Gutteridge alexg at ruggedtextile.com
Thu Sep 29 12:15:09 CEST 2011


 On Thu, 29 Sep 2011 10:07:06 +0200, Alyaa Mahmoud wrote:
> Dear BioC group
>
> I am using the collapseRows function from the WGCNA package, I need 
> to
> average expression data from probe sets that map to the same 
> EntrezID. Its
> stated that I have to idnetify two arguments rowID (corresponding to 
> probe
> set identifiers) and rowGroup (corresponding to the EntrezIDs)
>
> which I did below:
>
> summ=collapseRows(datET=ww, rowGroup=ww[,1], rowID=rownames(ww),
>                   method="Average")
>
> the output should be an expression matrix with expression data 
> averaged for
> multiple probes tha map to same geneID and rows corresponding to 
> geneIDs. My
> problem is in this later part, I don't get gene IDs in the rows, 
> instead
> this is what I get.
>
>    GSM97965.CEL.gz GSM97966.CEL.gz GSM97969.CEL.gz GSM97970.CEL.gz
> 1                2.176576        2.176576        2.176576        
> 2.176576
> 10               2.176576        2.176576        2.176576        
> 2.176576
> 100              4.517431        3.414851        3.225376        
> 3.113286
> 1000            12.330020       12.666929       10.414347       
> 10.479440
> 10000            7.051655        6.951304        7.111559        
> 8.566242
> 100009676        2.176576        2.176576        2.359885        
> 2.176576
>           GSM97971.CEL.gz GSM97972.CEL.gz
> 1                2.176576        2.176576
> 10               2.176576        2.176576
> 100              3.170813        3.093058
> 1000            10.300492       10.425190
> 10000            7.494686        7.415322
> 100009676        2.176576        2.176576
>
> any help would be much appreciated
> Thankss
> Alyaa

 Hi Alyaa,

 What do you mean by 'gene IDs in the rows'? It looks like the function 
 has worked fine to me. The rownames of the matrix you show look like 
 they are Entrez Gene IDs and the columns are the (I assume) mean 
 expression values for each gene.

-- 
 Alex Gutteridge



More information about the Bioconductor mailing list