[BioC] eBayes in LIMMA: borrowing which info?

Naomi Altman naomi at stat.psu.edu
Thu Mar 29 23:52:39 CEST 2007

>Q: is this only WHITHIN an array (i.e. columns), but also BETWEEN 
>arrays (i.e. rows)? In other words, the more arrays are present in 
>my eset, the better eBayes works?

The more replication within treatment, the smaller the effect of 
shrinkage.  This is because the error d.f. for the single gene error 
will increase, but the empirical prior d.f. will not change much.

>-->  I do have an experiment in which both wildtype and knockout 
>mice are analyzed; for now I am interested in the effects of a 
>nutrient in wildtype mice only. When I omit the KO mice, will that 
>influence the moderated t-test statistics for the wildtype mice 
>(putting aside normalization effects)?

Yes.  Both the d.f. and the estimated within error will change.

--Naomi Altman

>Thanks for now,
>Guido Hooiveld, PhD
>Nutrition, Metabolism & Genomics Group
>Division of Human Nutrition
>Wageningen University
>Biotechnion, Bomenweg 2
>NL-6703 HD Wageningen
>the Netherlands
>tel: (+)31 317 485788
>fax: (+)31 317 483342
>internet:   http://nutrigene.4t.com
>email:      guido.hooiveld at wur.nl
>         [[alternative HTML version deleted]]
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch
>Search the archives: 

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

More information about the Bioconductor mailing list