Hi Tarun,
I don't think you will like my answers...
"Tarun Nayar" <tnayar at="" bcgsc.ca=""> writes:
> 1. When I
>> ReadAffy(widget=T) for over about 20 HGU133aplus2 chips, I get the
> following message:
> Error: cannot allocate vector of size 232872 Kb
> In addition: Warning messages:
> 1: Reached total allocation of 1500Mb: see help(memory.size)
> 2: Reached total allocation of 1500Mb: see help(memory.size)
>
> I'm running Windows XP, 2Gb Ram, 3Ghz. Surely, I should be able to
> analyze more than 20 chips at a time with this system (?) I've
already
> upped my memory.limit...
It looks like R has indeed reached 1.5Gb and on a Windows system w/
2Gb RAM, I wouldn't expect to be able to access much more than that.
> Is there a cheaper way to create an Affy batch?
We are working on bounded memory algorithms, but they are not
available yet. Until then, obtaining a Linux system with lots of RAM
will be the most efficient approach.
> 2. When I want to update to the next version of R, how do I avoid
having
> to re-install all of the Bioconductor packages I already have
installed
> on my current version? At the moment, when I re-install, I can't
load
> any of the prior packages...
Because of changes in R, you will need to reinstall Bioconductor
packages.
Each Bioc release is targeted towards a particular R release. Bioc
1.7 goes with R 2.2.x. The upcoming Bioc 1.8 (to be released ~May 1)
is what you will want to use with the about to be released R-2.3.x
Best,
+ seth
Depending on what you want to do you may be able to skip the creation
of
the AffyBatch altogether. Specifically, if you are interested in RMA
or
GCRMA expression values you might get a little further down the track
using justRMA() or justGCRMA().
Ben
> (from a newbie...)
>
> 1. When I
> > ReadAffy(widget=T) for over about 20 HGU133aplus2 chips, I get the
> following message:
> Error: cannot allocate vector of size 232872 Kb
> In addition: Warning messages:
> 1: Reached total allocation of 1500Mb: see help(memory.size)
> 2: Reached total allocation of 1500Mb: see help(memory.size)
>
> I'm running Windows XP, 2Gb Ram, 3Ghz. Surely, I should be able to
> analyze more than 20 chips at a time with this system (?) I've
already
> upped my memory.limit...
> Is there a cheaper way to create an Affy batch?
Tarun, can you please send me the code that makes it crash + the
output of sessionInfo() + the operating system you are running on?
I need all this in order to be able to reproduce the crash on one
of our systems and try to sort out what's going on.
Thanks in advance,
H.
PS: Please, keep this discussion on the mailing list. Thanks!
Tarun Nayar wrote:
>Yes, it's crashing;
>
>After the memory limit warning comes up, the program crashes.
>
>-----Original Message-----
>From: Herve Pages [mailto:hpages at fhcrc.org]
>Sent: Wednesday, April 12, 2006 10:59 AM
>To: Tarun Nayar
>Subject: Re: [BioC] R crashing, updating
>
>Hi Tarun,
>
>In your subject, you are saying that R is crashing. Is it the case?
>Even if R can't allocate the requested memory, it shouldn't crash...
>
>H.
>
>
--
------------------------
Hervé Pagès
E-mail: hpages at fhcrc.org
Phone: (206) 667-5791
Fax: (206) 667-1319
