[BioC] ArrayQualityMetrics: Grey boxes in boxplot

[Audrey Kauffmann]

Dear Jeremy,

The grey boxes are the outliers as detected by the package (mean
and/or IQR different from the others).
They should correspond to the stars in the summary. Is that the case?
Which version of arrayQualityMetrics are you using?

Anyways, the outlier detection is here to help the user identifying
array(s) different from the others but the interpretation remains
yours. In this particular example, looking at the boxplots before and
after normalisation the arrays 1, 5, 8  and 9 seem fine to me.

Best wishes,
Audrey


2010/8/11 jeremy wilson <jeremy.wilson88 at gmail.com>:
> Dear all,
> I am noticing that some of my boxes in the boxplot are colored grey. I
> looked at the code to plot the boxplots (aqm.boxplot) and observed
> that there is a parameter "outlier" whose length if not zero  will
> color the respective arrays as grey. These outliers as I understand
> from the code are a union of outliers based on mean and variance of
> intensity distributions. Just by looking at the boxplots (attached), I
> do not think the 1st and the 5th arrays are outliers?
> The thing is that these were not shown as outliers in the report with
> a * mark in the before normalization report. The 9th array which is
> shown as grey in the after normalization plot was shown as outlier (*)
> when the eset after normalization was used to perform AQM.
> Are the grey colored ones really outliers or some thing else? If they
> are outliers, why are the three arrays from the before normalization
> boxplot not marked as outliers in the report?
>
> Please let me know
> Thank you
>
> _______________________________________________
> 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
>



-- 
Audrey Kauffmann
Bergonie Cancer Institute
229 Cours de l'Argonne
33076 Bordeaux
France