[R] How to check cy5 and cy3 values were lowess normalized

[Henrik Bengtsson]

On Thu, Feb 14, 2008 at 4:57 AM, Ng Stanley <stanleyngkl at gmail.com> wrote:
> Hi,
>
>  I have some microarray data,  cy5 and cy3 values are in log2. Is there a
>  way to check they have undergone lowess normalization ?

Yes, go back ask the one who you got the data from.

Honestly, this is a serious reply, because why waste time second
guessing what data you are working with.  I've seen people (including
me) spending way too much time to knowing exactly what data I am
working with.  It is just a waste of time and in the end of the day it
should be possible to find out from the source.

To answer your question as is, if you know that the data is *either*
curve-fit ("lowess" is one of them) normalized *or* not at all, then a
log-ratio log-intensity scatter plot *might* show symptoms that are
very specific to curve-fit normalized data.  These symptoms are
stronger the more curvature there was in the log-ratio log-intensity
plot for the beginning.  The best way I illustrate this symptom is to
look at the symmetry of the normalized data down at the lower
intensities.  See Panel "B (curve-fit)" on Slide 16 in

 http://www.stat.berkeley.edu/hb/talks/BengtssonH_20080208-YetAnotherAffinePresentation.pdf

Going back to the (R,G) space on intensity scale will also reveal the
same symptoms, cf. Slide 11.

You'll find all the background in:

H. Bengtsson and O. Hössjer, Methodological study of affine
transformations of gene expression data with proposed robust
non-parametric multi-dimensional normalization method, BMC
Bioinformatics, 2006, 7:100.

Finally, you reach more bioinformaticians/statisticians in the field
if you send these questions to the http://www.bioconductor.org/
mailing list instead.  Please continue the thread there if you have
follow-up questions.

Cheers

/Henrik

>
>  Thanks
>  Stanley
>
>         [[alternative HTML version deleted]]
>
>  ______________________________________________
>  R-help at r-project.org mailing list
>  https://stat.ethz.ch/mailman/listinfo/r-help
>  PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>  and provide commented, minimal, self-contained, reproducible code.
>