[BioC] chip area distortion tolerance

James W. MacDonald jmacdon at med.umich.edu
Tue Dec 12 15:41:03 CET 2006


Hi Anna,

Please don't take things off-list.

Thalacker-Mercer, Anna E wrote:
> Thank you for your response.  There were a few RLE and NUSE plots that
> did not look so good and I think those chips should be removed.  
> 
> If I could ask you one more question.  I ran the PLM analysis on all 66
> chips and then on the 6 individual treatments.  The results of the
> latter RLE plots were slightly modified (i.e. a few chips were no longer
> problematic when I looked at the individual treatments).  I haven't been
> able to find anywhere which way is better to run the PLM analyses, but I
> have seen both.  Any thoughts?  

I am not sure what your experiment entails, so don't know exactly what 
you mean by individual treatments. My best guess is that you have 
samples that were treated six different ways, and you want to compare 
the effect of the treatments on the gene expression.

If that is the case, then you almost certainly want to compute 
expression values as a group rather than on each individual treatment. I 
can't think of any reason for doing things on each treatment, and I can 
think of many reasons not to.

Best,

Jim


> 
> Thank you again for your help,
> 
> Best regards,
> Anna
> 
> -----Original Message-----
> From: James W. MacDonald [mailto:jmacdon at med.umich.edu] 
> Sent: Tuesday, December 12, 2006 9:20 AM
> To: Thalacker-Mercer, Anna E
> Cc: bioconductor at stat.math.ethz.ch
> Subject: Re: [BioC] chip area distortion tolerance
> 
> Thalacker-Mercer, Anna E wrote:
> 
>>Hello,
>>I have a question in addition to your question about distortions.  How
>>much should artifacts influence taking the chip out of the analysis?
> 
> I
> 
>>have several chips that have decent sized artifacts in them, but seem
>>fine when looking at the RLE and NUSE plots.  
> 
> 
> If the QA plots look OK, then you can probably use the chips if you are 
> using a robust model fit like RMA or GCRMA. Remember that the probesets 
> are distributed all over the chip, so an artifact on one section of the 
> chip will probably not have a large effect on any one probeset.
> 
> Best,
> 
> Jim
> 
> 
> 
>>Thanks
>>Anna
>>
>>-----Original Message-----
>>From: bioconductor-bounces at stat.math.ethz.ch
>>[mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of D.Enrique
>>ESCOBAR ESPINOZA
>>Sent: Monday, December 11, 2006 11:27 AM
>>To: bioconductor at stat.math.ethz.ch
>>Subject: [BioC] chip area distortion tolerance
>>
>>Hi,
>>I am doing some QC for chips,
>>in the article of HEBER & SICK (2006),
>>they use a threshold of 20% of distortion of the area chip.
>>
>>!) is there any consensus?
>>2) how do you estimate this %percentage of distortion in R?
>>
>>_______________________________________________
>>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
>>
>>_______________________________________________
>>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
> 
> 


-- 
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.



More information about the Bioconductor mailing list