Entering edit mode
Until proven otherwise, I think that a channel by channel analysis in
which
array is entered (preferably as a random effect) along with dye is
valid. I.e. I treat the entire design as an ANOVA and look at any
contrast
that is reasonable.
--Naomi
At 03:09 PM 9/14/2004 +0100, michael watson (IAH-C) wrote:
>Hi
>
>I wondered what the list felt about calculating ratios between
channels
>from separate microarrays?
>
>Let me explain. I have come across an experiment (not designed by
>myself) where I have the following:
>
>+ve cell lines - cell lines with random genomic insertions that
display
>a +ve phenotype
>-ve cell lines - cell lines with random genomic insertions that
display
>a -ve phenotype
>Parental cell lines with no genomic insertion
>
>Now to me it would have made sense to compare +ve vs parental and -ve
vs
>parental to find those genes which are present in the insertions but
not
>in the parent which may be contributing to the phenotype.
>
>However, the experimenter did not do this; they compared each +ve
cell
>line to the -ve cell lines, and also the parental cell lines to the
-ve
>cell lines.
>
>So intuitively what I want to do is calculate ratios of +ve/parental
and
>-ve/parental for each of the experiments, when what I have is +ve/-ve
>and parental/-ve.
>
>Is it valid simply to take the (single) list of values for the
parental
>cell line and use that repeatedly as the denominator for the +ve and
-ve
>channels from the other arrays?
>
>Thanks
>Mick
>
>_______________________________________________
>Bioconductor mailing list
>Bioconductor@stat.math.ethz.ch
>https://stat.ethz.ch/mailman/listinfo/bioconductor
Naomi S. Altman 814-865-3791 (voice)
Associate Professor
Bioinformatics Consulting Center
Dept. of Statistics 814-863-7114 (fax)
Penn State University 814-865-1348
(Statistics)
University Park, PA 16802-2111