Dear List
I'm new to microarrays analysis. I'm analyzing an experiment with 2
condtions and 3 microarrays for condition. My data are succesfully
normalized and I have a list of gene modulated with aFC, a raw p-value
and a
q-value. My q-value are never lower than 0,1. If I well understand my
q-value is the minimum false discovery rate for which my FC are
statistically significant. So have I set a threshold similar to
p-value to
select my differentially regulated genes? if yes (what does mean the
fact
that my q-value is never lower than 0,1?)
Sorry for my confused question and
Thank you for every tips you could give to me
--
Guido Leoni
National Research Institute on Food and Nutrition
(I.N.R.A.N.)
via Ardeatina 546
00178 Rome
Italy
tel + 39 06 51 49 41 (operator)
+ 39 06 51 49 4519 (direct)
[[alternative HTML version deleted]]
Hi Guido,
Guido Leoni wrote:
> Dear List
> I'm new to microarrays analysis. I'm analyzing an experiment with 2
> condtions and 3 microarrays for condition. My data are succesfully
> normalized and I have a list of gene modulated with aFC, a raw
p-value and a
> q-value. My q-value are never lower than 0,1. If I well understand
my
> q-value is the minimum false discovery rate for which my FC are
> statistically significant. So have I set a threshold similar to
p-value to
> select my differentially regulated genes? if yes (what does mean the
fact
> that my q-value is never lower than 0,1?)
> Sorry for my confused question and
> Thank you for every tips you could give to me
The FDR is an estimate of the maximum percentage of 'significant'
results that are false positives. So for instance, if you accept all
the
genes (reporters) with a q-value =< 0.1, then the assumption is that
somewhere around 10% of those are false positives.
So when you go to validate those genes (if you use that FDR level),
you
can expect about 10% or so to fail to validate. This of course assumes
that all the assumptions have been met blah blah statistics statistics
statistics... ;-D
Best,
Jim
>
--
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should
not be used for urgent or sensitive issues
Dear James, Guido and Bioconductor helpers,
I always thought, q-values control pFDR, which is similar to FDR
almost all
the time but theoretically not similar.
Please correct me if I am wrong.
Thanks and Best Regards,
Sanvesh
On Wed, Feb 3, 2010 at 5:14 PM, James W. MacDonald
<jmacdon@med.umich.edu>wrote:
> Hi Guido,
>
>
> Guido Leoni wrote:
>
>> Dear List
>> I'm new to microarrays analysis. I'm analyzing an experiment with 2
>> condtions and 3 microarrays for condition. My data are succesfully
>> normalized and I have a list of gene modulated with aFC, a raw
p-value and
>> a
>> q-value. My q-value are never lower than 0,1. If I well understand
my
>> q-value is the minimum false discovery rate for which my FC are
>> statistically significant. So have I set a threshold similar to
p-value to
>> select my differentially regulated genes? if yes (what does mean
the fact
>> that my q-value is never lower than 0,1?)
>> Sorry for my confused question and
>> Thank you for every tips you could give to me
>>
>
> The FDR is an estimate of the maximum percentage of 'significant'
results
> that are false positives. So for instance, if you accept all the
genes
> (reporters) with a q-value =< 0.1, then the assumption is that
somewhere
> around 10% of those are false positives.
>
> So when you go to validate those genes (if you use that FDR level),
you can
> expect about 10% or so to fail to validate. This of course assumes
that all
> the assumptions have been met blah blah statistics statistics
statistics...
> ;-D
>
> Best,
>
> Jim
>
>
>
>>
> --
> James W. MacDonald, M.S.
> Biostatistician
> Douglas Lab
> University of Michigan
> Department of Human Genetics
> 5912 Buhl
> 1241 E. Catherine St.
> Ann Arbor MI 48109-5618
> 734-615-7826
> **********************************************************
> Electronic Mail is not secure, may not be read every day, and should
not be
> used for urgent or sensitive issues
> _______________________________________________
> Bioconductor mailing list
> Bioconductor@stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
[[alternative HTML version deleted]]
