Dear Bioconductors,
I apologise if this is not 100% appropriate for the list. I have seen
data sets from a couple of experiments lately, each using four MOE430
version 2 chips, in which one chip of the four has a median PM which
is a
factor of three or four higher than on the others. When I do an RNA
degradation plot, the aberrant chip has a big dip at the 3' end (the
others look fine - the mean log PM increases pretty much linearly as
you
go 5' -> 3', and they are roughly parallel to each other) . Is this
really indicative of RNA degradation, or am I misinterpreting or
reading
too much into the plot? It seems counterintuitive that a degraded
sample
would produce *higher* intensities. My collaborators did check their
samples before labelling (using a Bioanalyser) and they all appeared
to be
fine, so the cause of the problem would appear to be downstream. If
anyone else has seen similar behaviour, I'd be most interested to know
-
especially if they worked out what the cause was!
Thanks in advance,
Ken
--
Dr. Ken Simpson
Research Officer, Division of Genetics and Bioinformatics
The Walter and Eliza Hall Institute of Medical Research
1G Royal Parade, Parkville, Vic 3050
Tel: (03) 9345 2628
---
Ken Simpson wrote:
> Dear Bioconductors,
>
> I apologise if this is not 100% appropriate for the list. I have
seen
> data sets from a couple of experiments lately, each using four
MOE430
> version 2 chips, in which one chip of the four has a median PM which
is a
> factor of three or four higher than on the others. When I do an RNA
> degradation plot, the aberrant chip has a big dip at the 3' end (the
> others look fine - the mean log PM increases pretty much linearly as
you
> go 5' -> 3', and they are roughly parallel to each other) . Is this
> really indicative of RNA degradation, or am I misinterpreting or
reading
> too much into the plot? It seems counterintuitive that a degraded
sample
> would produce *higher* intensities. My collaborators did check
their
> samples before labelling (using a Bioanalyser) and they all appeared
to be
> fine, so the cause of the problem would appear to be downstream. If
> anyone else has seen similar behaviour, I'd be most interested to
know -
> especially if they worked out what the cause was!
We have not worked out the cause for this sort of thing, but in our
experience this is usually caused by something going awry in the
fragmentation step. Whenever we see this sort of thing, we re-run that
sample starting at the fragmentation step and in probably 90% of the
cases that is all you need to do. Unfortunately, the majority of the
cost of an Affy experiment is the chip, so it is an expensive fix...
Jim
>
> Thanks in advance,
>
> Ken
>
--
James W. MacDonald
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623
