[BioC] out of memory problem using affycoretools

[Al Ivens]

Hi,

Have you tried Ben Bolstad's RMAExpress do the initial bg and
normalisation?  This stand-alone works well for big datasets.  A Windows
exe is available:

http://rmaexpress.bmbolstad.com/

a

> -----Original Message-----
> From: bioconductor-bounces at stat.math.ethz.ch 
> [mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of 
> James Anderson
> Sent: 07 December 2006 14:56
> To: bioconductor
> Subject: [BioC] out of memory problem using affycoretools
> 
> 
> Hi, 
>   I have about 53 cel files, each one has the size of about 
> 12-13 mega bytes with 22625 probes. I am using affycoretools 
> to do rma, but it complains that the maximum size of memory 
> has been reached. My PC is Windows XP with 2G RAM. Below is 
> the error information. 
> 
> Error: cannot allocate vector of size 209906 Kb
> In addition: Warning messages:
> 1: Incompatible phenoData object. Created a new one.
>  in: read.affybatch(filenames = filenames, phenoData = phenoData) 
> 2: Reached total allocation of 1024Mb: see help(memory.size) 
> 3: Reached total allocation of 1024Mb: see help(memory.size) 
> 
> Since RMA is an across sample normalization, and I need to 
> analyze all the 53 samples together, I can not separate them 
> into several pieces and then do RMA, MAS5 can do that, but 
> the expression value obtained by MAS5 is not as good as that 
> of RMA. Does anybody know how to solve this problem? In 
> addition, it seems that the output of rma in affy package 
> (not affycoretools) has several lines for each individual 
> probe (my case is 11 lines for each individual probe), while 
> the rma in affycoretools directly gives the final expression 
> value for each probe, could anybody tell me how to make the 
> two results from affy and affycoretools comparable with each other? 
> 
> Thanks a lot!
> 
> James
> 
>  
> ---------------------------------
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch 
> https://stat.ethz.ch/mailman/listinfo/biocondu> ctor
> Search the 
> archives: 
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>