Hi Sean,
Thanks for the reply. Upon closer inspection of the data, it appears
that
the qvalue package is working, but the output is just wonky. I get
numbers
that are the same value near 1 for all genes for most of my
comparisons. The
B-statistic is also a negative value. Could there be something in the
gpr
file that is causing this to occur?
Darren
-----Original Message-----
From: Sean Davis [mailto:sdavis2@mail.nih.gov]
Sent: Thursday, June 01, 2006 4:01 AM
To: Darren Ames
Subject: Re: [BioC] flat microarray p-values
On 5/31/06 6:23 PM, "Darren Ames" <ames at="" ag.arizona.edu=""> wrote:
> I have a problem with my latest batch of microarrays. A histogram of
all
the
> p-values is flat, so the qvalue package does not work. Quality
control
> indicates that these are good arrays. Does anyone have any ideas?
What doesn't work about the qvalue package? (I don't use it much, so
I may
not be helpful, but....) What commands have you tried and what is the
output?
Sean
On 6/1/06 3:02 PM, "Darren Ames" <ames at="" ag.arizona.edu=""> wrote:
> Hi Sean,
>
> Thanks for the reply. Upon closer inspection of the data, it appears
that
> the qvalue package is working, but the output is just wonky. I get
numbers
> that are the same value near 1 for all genes for most of my
comparisons. The
> B-statistic is also a negative value. Could there be something in
the gpr
> file that is causing this to occur?
All of this points to a lack of evidence for differential expression.
Perhaps n is too small (lack of power) or there is no differential
expression between groups.
Sean
