[BioC] memory to use affy package

Stephen Henderson s.henderson@ucl.ac.uk
Sat, 7 Sep 2002 16:58:59 +0100


My computer has 384Mb

I'm told 256 should be sufficient (if slow). I think I have a problem with
the memory allocation of my computer to R or my understanding of said
process.

I'll try reloading R--and see what happens.


-----Original Message-----
From: Laurent Gautier [mailto:laurent@cbs.dtu.dk] 
Sent: Friday, September 06, 2002 6:29 PM
To: Stephen Henderson
Cc: 'bioconductor@stat.math.ethz.ch'
Subject: Re: [BioC] memory to use affy package

On Fri, Sep 06, 2002 at 04:20:52PM +0100, Stephen Henderson wrote:
> I have been trying to use the affy package. I am running the Win32 version
> but I am finding that calculating expression values (HG133A X 6) crashes
R.
> that is I run out of memory. 
> 
> e.g
> > express(Data,normalize=T, normalize.method="quantiles")
> Normalizing Data
> Background correcting
> Preparing Data
> Error: cannot allocate vector of size 32768 Kb
> In addition: Warning message: 
> Reached total allocation of 500Mb: see help(memory.size)
> 
> I have allocated the maximum for this computer 
> 
> e.g.
> > memory.limit(whatever)
> 
> I have 3 questions that I would be grateful if anyone could help me with:-
> 
> 1. Do all expression calculation and normalisation methods use similar
> amounts of memory (e.g loess?)?

No.

> 
> 2. Does the memory requirement for some methods rise as the chip numbers
> rise?

Yes. (for all methods)

> 
> 3. Reasonably how much memory do I need on a WIN32 to analyse HG133A
chips?
> 

It depends on the number of chips. I am not very knowledgeable about 
the behavior of R and the package under Windows. Could you tell more
about the number of chips you have loaded and the total amount of
RAM your computer has ?

> Many Thanks
> Stephen Henderson
> 
> 
> 
> **********************************************************************
> This email and any files transmitted with it are confidential and
> intended solely for the use of the individual or entity to whom they
> are addressed. If you have received this email in error please notify
> the system manager (rmgztem@ucl.ac.uk). All files are scanned for viruses.
> **********************************************************************
> 

-- 
--------------------------------------------------------------
Laurent Gautier			CBS, Building 208, DTU
PhD. Student			DK-2800 Lyngby,Denmark	
tel: +45 45 25 24 89		http://www.cbs.dtu.dk/laurent