[BioC] Accessing SD values from CEL files

[James W. MacDonald]

Hi Guido,

Hooiveld, Guido wrote:
> Hi,
> 
> To follow-up on this question:
> What exactely do these se (and exprs) values represent?

The exprs slot holds the intensity value reported by the Affy software 
for each probe. The se.exprs slot holds the standard deviation values 
reported by the Affy software.

The intensity value reported by the Affy software is the 75th percentile 
of the pixels that remain when the outermost pixels are discarded (for 
the current 3' arrays this is about 16 pixels). I would imagine the 
stdev is calculated on these 16 pixels as well.

Best,

Jim



> 
> I am asking because when I list the 'content' of the expres and se of a
> ReadAffy(sd=TRUE) object, these contain many more entries than the
> number of probesets (~950,000 vs 45,101). So these are likely thus the
> intesnities of individual probes. But how then to calculate the SD from
> an individual probe value? (I understand this can be calculated for a
> probeset, but not for a probe). 
> 
> My session for 2 arrays:
> 
> 
>>data <- ReadAffy(sd=TRUE)
>>data
> 
> AffyBatch object
> size of arrays=1002x1002 features (31382 kb)
> cdf=Mouse430_2 (45101 affyids)
> number of samples=2
> number of genes=45101
> annotation=mouse4302
> 
> 
>>exprs <- exprs(data)
>>exprs
> 
>   [49984,]                               91
> 85
>   [49985,]                              228
> 165
>   [49986,]                              194
> 142
>   [49987,]                              438
> 335
>   [49988,]                              184
> 136
>   [49989,]                              122
> 91
>   [49990,]                               58
> 44
>   [49991,]                              243
> 219
>   [49992,]                              118
> 78
>   [49993,]                               54
> 47
>   [49994,]                               75
> 51
>   [49995,]                               61
> 48
>   [49996,]                              131
> 141
>   [49997,]                               63
> 50
>   [49998,]                              128
> 105
>   [49999,]                               77
> 59
>  [ reached getOption("max.print") -- omitted 954005 rows ]]
> 
> 
>>se <- se.exprs(data)
>>se
> 
>   [49984,]
> 13.2                                                                8.5
>   [49985,]
> 25.3                                                               28.6
>   [49986,]
> 27.3                                                               22.8
>   [49987,]
> 44.5                                                               48.8
>   [49988,]
> 32.8                                                               23.6
>   [49989,]
> 27.1                                                               15.2
>   [49990,]
> 10.9                                                                7.7
>   [49991,]
> 49.6                                                               35.9
>   [49992,]
> 15.1                                                               13.6
>   [49993,]
> 7.6                                                                7.3
>   [49994,]
> 11.1                                                                7.8
>   [49995,]
> 7.6                                                                6.2
>   [49996,]
> 24.4                                                               21.8
>   [49997,]
> 9.5                                                                9.6
>   [49998,]
> 24.7                                                               17.5
>   [49999,]
> 12.5                                                               11.2
>  [ reached getOption("max.print") -- omitted 954005 rows ]]
> 
> 
> Thanks for enlighten me,
> 
> Guido
> 
>  
> ------------------------------------------------
> Guido Hooiveld, PhD
> Nutrition, Metabolism & Genomics Group
> Division of Human Nutrition
> Wageningen University
> Biotechnion, Bomenweg 2
> NL-6703 HD Wageningen
> the Netherlands
>  
> tel: (+)31 317 485788
> fax: (+)31 317 483342
>  
> internet:   http://nutrigene.4t.com
> email:      guido.hooiveld at wur.nl
> 
> 
> 
>>Date: Mon, 02 Apr 2007 16:21:32 -0400
>>From: "James W. MacDonald" <jmacdon at med.umich.edu>
>>Subject: Re: [BioC] Accessing SD values from CEL files
>>To: hvanleeuwen at ucdavis.edu
>>Cc: bioconductor at stat.math.ethz.ch
>>Message-ID: <461165CC.5000203 at med.umich.edu>
>>Content-Type: text/plain;  charset="utf-8";  format=flowed
>>
>>Hi Hans,
>>
>>Hans van Leeuwen wrote:
>>
>>> 
>>>Hi,
>>>
>>>I want to extract both the signal intensity and the 
>>
>>corresponding SD values
>>
>>>from Affymetrix CEL files. I read in the affy BioC 1.8 
>>
>>release notes that
>>
>>>ReadAffy can now optionally return the SD values stored in 
>>
>>the CEL file.
>>
>>>When I use ReadAffy I add the parameter sd = TRUE, but I am 
>>
>>trying to figure
>>
>>>out how I can get these value from the object. Is there a function,
>>>equivalent to the intensities() function, to retrieve the SD values?
>>
>>Yes. The function se.exprs() will do this. See 
>>?AffyBatch-class for more 
>>information.
>>
>>Best,
>>
>>Jim
>>


-- 
James W. MacDonald, M.S.
Biostatistician
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623


**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.