[BioC] How does one deal with spatial effects detected on single channel microarrays?

Mon Sep 2 20:13:19 CEST 2013

Dear Scott,

I forgot to mention one QC, namely the density plots of the raw data. If 
some of the mentioned QCs show that the corresponding CEL-file may have 
problems, I would simply remove it.

Best regards,
Christian

On 9/2/13 7:18 PM, Scott Robinson wrote:
> Dear Christian,
>
> Brilliant, thanks very much. I have just one last query: if I were to have a more detrimental spatial effect do you (or anyone) know what BioC package (if any exists) would be appropriate for spatial normalization of single-channel Affy chips? Or is it simply a case of removing that sample?
>
> Many thanks,
>
> Scott
>
> PS I posted this same question a week or two ago on biostars.org but got no answers. I hope you wouldn't mind if I formulate an 'answer' out of what you have said to post on biostars, and accrediting you?
>
> -----Original Message-----
> From: cstrato [mailto:cstrato at aon.at]
> Sent: 02 September 2013 18:09
> To: Scott Robinson
> Cc: Scott Robinson [guest]; bioconductor at r-project.org
> Subject: Re: [BioC] How does one deal with spatial effects detected on single channel microarrays?
>
> Dear Scott,
>
> Yes and yes, however even the crop circles may be ok. There are a lot of quality controls that you can do additionally, e.g. PCA, NUSE, RLE, border plots, center of intensity plots, etc.
>
> Best regards,
> Christian
>
>
> On 9/2/13 6:41 PM, Scott Robinson wrote:
>> Dear Christian,
>>
>> Thanks very much for the help. I especially like the "crop circles" artefact.
>>
>> So is the idea that if you have a small spatial artefact it's probably going to affect only a small number of the probes in each probe set and therefore not affect the summarised values much? Do you only have to spatially normalize or remove a chip from analysis if the spatial artefact is quite large? Maybe if it covers 1/4 of the chip or something?
>>
>> Thanks,
>>
>> Scott
>>
>> -----Original Message-----
>> From: cstrato [mailto:cstrato at aon.at]
>> Sent: 02 September 2013 16:28
>> To: Scott Robinson [guest]
>> Cc: bioconductor at r-project.org; Scott Robinson
>> Subject: Re: [BioC] How does one deal with spatial effects detected on single channel microarrays?
>>
>> Dear Scott,
>>
>> Unlike cDNA arrays Affymetrix arrays use between 11 and 20 oligonucleotides per transcript. These oligos were placed in one line on the first Hu6800 array, but since a long time these oligos are scattered randomly across the whole array, in order to prevent spatial effects.
>>
>> The images of your arrays are all ok.
>> You can see some weird examples at:
>> http://plmimagegallery.bmbolstad.com/
>>
>> Best regards
>> Christian
>> _._._._._._._._._._._._._._._._._._
>> C.h.r.i.s.t.i.a.n   S.t.r.a.t.o.w.a
>> V.i.e.n.n.a           A.u.s.t.r.i.a
>> e.m.a.i.l:        cstrato at aon.at
>> _._._._._._._._._._._._._._._._._._
>>
>>
>>
>> On 9/2/13 2:44 PM, Scott Robinson [guest] wrote:
>>>
>>> Dear all,
>>>
>>> I have been doing pre-processing & QC of a number of CEL files (from Affymetrix U133+ v2.0 chips), basing things loosely on this tutorial (http://bioinformatics.knowledgeblog.org/2011/06/20/analysing-microarray-data-in-bioconductor/), and on some past experience of other microarray technologies.
>>>
>>> Many tutorials seem to deal with indentification of spatial effects but do not discuss how to handle them. As such I have been having difficulty finding methods directed at spatial normalization in single-channel arrays (OLIN, smida, marray and nnNorm all seem to be written for dual-channel arrays). Can anyone please suggest an appropriate package/method for single-channel Affy chips?
>>>
>>> And are there cases where they should be excluded rather than normalized?
>>>
>>> Some examples of my spatial artefacts:
>>>
>>> http://postimg.org/image/kkfjt4o1j/
>>> http://postimg.org/image/v5utre4zb/
>>> http://postimg.org/image/yfdublign/
>>>
>>> Of my 90 chips example 2 is the only one of that kind of pattern. The rest are mostly similar in form to the other 2 and similar patterns are seen in ~20 chips.
>>>
>>> Thanks in advance,
>>>
>>> Scott
>>>
>>>     -- output of sessionInfo():
>>>
>>> R version 3.0.1 (2013-05-16)
>>> Platform: x86_64-w64-mingw32/x64 (64-bit)
>>>
>>> locale:
>>> [1] LC_COLLATE=English_United Kingdom.1252 [2]
>>> LC_CTYPE=English_United
>>> Kingdom.1252 [3] LC_MONETARY=English_United Kingdom.1252 [4]
>>> LC_NUMERIC=C [5] LC_TIME=English_United Kingdom.1252
>>>
>>> attached base packages:
>>> [1] parallel  stats     graphics  grDevices utils     datasets  methods
>>> [8] base
>>>
>>> other attached packages:
>>>     [1] limma_3.16.7          sparcl_1.0.3          lattice_0.20-23
>>>     [4] corrplot_0.71         affyPLM_1.36.0        preprocessCore_1.22.0
>>>     [7] simpleaffy_2.36.1     gcrma_2.32.0          genefilter_1.42.0
>>> [10] affy_1.38.1           Biobase_2.20.1        BiocGenerics_0.6.0
>>>
>>> loaded via a namespace (and not attached):
>>>     [1] affyio_1.28.0        annotate_1.38.0      AnnotationDbi_1.22.6
>>>     [4] BiocInstaller_1.10.3 Biostrings_2.28.0    DBI_0.2-7
>>>     [7] grid_3.0.1           IRanges_1.18.3       RSQLite_0.11.4
>>> [10] splines_3.0.1        stats4_3.0.1         survival_2.37-4
>>> [13] XML_3.98-1.1         xtable_1.7-1         zlibbioc_1.6.0
>>>
>>> --
>>> Sent via the guest posting facility at bioconductor.org.
>>>
>>> _______________________________________________
>>> Bioconductor mailing list
>>> Bioconductor at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>> Search the archives:
>>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>>
>>
>