[BioC] extracting significant genes using limma
James W. MacDonald
jmacdon at med.umich.edu
Tue Mar 14 15:41:58 CET 2006
Naomi Altman wrote:
> Dear Assa,
> This one will need to go to the list. I only use topTable.
> p.s. Please post all questions directly to the listserv. That way
> you can get more help than any one person can give, and no-one is
> At 03:29 AM 3/14/2006, you wrote:
>>I was wondering if I can use decideTests()-function in my tests,
>>even if i didn't use any contrast matrix.
Yes you can. However, depending on how you set up your design matrix you
may or may not get the comparisons you are looking for.
>>I didn't get the difference between decideTests() and classifyTests().
decideTests() is a wrapper function that calls classifyTestsP() or
classifyTestsF() if you select method = "heirarchical" or "nestedF",
respectively. decideTests() also uses different default p-value cutoffs
and multiplicity corrections. Looking at the code for decideTests() may
help you understand the differences.
>>I would be happy for any help you can give me.
>>On 3/13/06, Naomi Altman
>><<mailto:naomi at stat.psu.edu>naomi at stat.psu.edu> wrote:
>>Since you used "adjust=fdr", the p-value column of the TopTable are
>>the "adjusted p-values" after fdr (which I think of as q-values).
>>You can either pick some q-value you want to use to select the
>>significantly differentially expressing genes, or you can pick some
>>number of genes, and report the q-value of the least significant of these.
>>At 10:17 AM 3/13/2006, Assa Yeroslaviz wrote:
>>>I know this theme is an old one, but I look all over the archives and didn't
>>>find any help regarding this subject.
>>>Using Affymetrix chips I compared two groups (Control vs compound) with the
>>>I made an affybatch Object using ReadAffy(), normalised the data with the
>>>RMA algorithm and fitted a linear model with lmFit.
>>>>affy <- ReadAffy(filenames=vec)
>>>>eset <- rma(affy)
>>>>design <- cbind(Control=1,AE0627vsCT=c(rep(0,6),rep(1,4)))
>>>my design matrix looks like that (I have 6 control and 4 treated arrays):
>>> Control AE143vsCT
>>> [1,] 1 0
>>> [2,] 1 0
>>> [3,] 1 0
>>> [4,] 1 0
>>> [5,] 1 0
>>> [6,] 1 0
>>> [7,] 1 1
>>> [8,] 1 1
>>> [9,] 1 1
>>>[10,] 1 1
>>>so I don't need any contrast matrix.
>>>The list is 22,810 genes long. But not all of them can be significant. I
>>>I sorted the genes with:
>>>>sig_table <- topTable(fit_e, coef=2, number=6000, adjust="fdr", sort.by=
>>>I've chosen 6000 as an arbitrary value, but I still don't know how many
>>>genes are siginificant.
>>>My question(s) is(are):
>>>1. How do I find out how many genes are significantly differentially
>>>expressed using a define p-value or FDR?
>>> Can I use here the decideTests() function although I don't have any
>>>2. In SAM one can look for the false discovery rates using the different
>>> Is it possible to set a fixed FDR-Value in Limma?
>>> Where Do I find the FDR rates of my significant genes?
>>>3. Is there a possibility (like in SAM) to show the results in a graphic (
>>>scatter plot etc.)?
>>>Every comment and suggestion would be appreciated!
>>>Loetzener Str. 15
>>> [[alternative HTML version deleted]]
>>>Bioconductor mailing list
>>><mailto:Bioconductor at stat.math.ethz.ch>Bioconductor at stat.math.ethz.ch
>>Naomi S. Altman 814-865-3791 (voice)
>>Dept. of Statistics 814-863-7114 (fax)
>>Penn State University 814-865-1348 (Statistics)
>>University Park, PA 16802-2111
>>Loetzener Str. 15
> Naomi S. Altman 814-865-3791 (voice)
> Associate Professor
> Dept. of Statistics 814-863-7114 (fax)
> Penn State University 814-865-1348 (Statistics)
> University Park, PA 16802-2111
> [[alternative HTML version deleted]]
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
James W. MacDonald, M.S.
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
Ann Arbor MI 48109
More information about the Bioconductor