[BioC] running out of memory

[Javier Pérez Florido]

Yep, at the normalization step.
The size of CEL files are around 800MB in total,
Javier


Carlos J. Gil Bellosta escribió:
> Hello,
>
> How big are your 100 CEL files in total? At what point does it break?
> Just at the normalization step?
>
> Regards,
>
> Carlos J. Gil Bellosta
> http://www.datanalytics.com
>
> El día 20 de enero de 2010 17:24, Javier Pérez Florido
> <jpflorido at gmail.com> escribió:
>   
>> Dear Steve,
>> Thanks for your help, but, I'm trying to preprocess using RMA, GCRMA, VSN
>> and dChip methods and I don't see any function in aroma.affymetrix that
>> performs the whole preprocessing...
>>
>>
>>
>> Steve Lianoglou escribió:
>>     
>>> Hi,
>>>
>>> 2010/1/20 Javier Pérez Florido <jpflorido at gmail.com>:
>>>
>>>       
>>>> Dear list,
>>>> I'm trying to normalize using threestep several CEL files (around 100
>>>> CEL files). I got the following error:
>>>>
>>>> Error: cannot allocate vector of size 6.5 Mb
>>>>
>>>> I've been reading the suggestions about this error in the mailing list,
>>>> but I couldn't fix it. My system (Win XP Prof, 32 bits) has 3 GB RAM and
>>>> I did the following:
>>>>
>>>>   * I used the command line --max-mem-size=3071M in the R shortcut
>>>>   * I changed the boot.ini file to allow up to 3GB
>>>>         o [boot loader]
>>>>           timeout=30
>>>>           default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
>>>>           [operating systems]
>>>>           multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Microsoft
>>>>           Windows XP Professional" /3GB /noexecute=optin /fastdetect
>>>>   * I try to remove not used variables using the rm() command and the
>>>>     garbage collector gc() within the source code.
>>>>
>>>> But, even with those changes, I'm still running out of memory. It is
>>>> strange, because, having a look at the tasks administrator, R doesn't
>>>> use more than 1600 MB and the whole system never goes further than 2GB
>>>>
>>>>         
>>> I can't really comment on tweaking memory settings on windows systems,
>>> but if all you're doing is trying to normalize a boat-load of affy
>>> arrays together, I understand that the aroma.affymetrix package can do
>>> so while keeping memory requirements down. Perhaps you might consider
>>> looking into it until someone can give you better advice:
>>>
>>> http://groups.google.com/group/aroma-affymetrix/web/overview
>>>
>>> HTH,
>>> -steve
>>>
>>>
>>>       
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>>     
>
>