[BioC] Difference between LIMMA and Genespring

[Gordon K Smyth]

Dear Ming,

Limma borrows information between genes, whereas Genespring does not. 
This gives limma more statistical power to detect more differentially 
expressed genes, and also produces as more reliable gene ranking with 
fewer false discoveries.  When the sample sizes are small, the differences 
can be dramatic.

See Shirley Liu's primer published in Nature Genetics for a tutorial on 
what limma does and why it works:

  http://www.nature.com/nbt/journal/v28/n4/abs/nbt.1619.html

For reasons that I don't understand, Genespring has chosen not to 
implement any of the statistical methods developed in the past decade or 
so for differential expression of microarrays.  They stick to t-tests and 
anova methods from early last century, methods that are unsuitable for 
genomic data with small samples sizes.

As has been remarked by another poster, there may be other differences as 
well between Genespring as limma as to do with pre-processing.

Best wishes
Gordon

> Date: Mon, 12 Dec 2011 14:42:38 +0100
> From: "Jin, Ming" <ming.jin at helmholtz-muenchen.de>
> To: "'bioconductor at r-project.org'" <bioconductor at r-project.org>
> Subject: [BioC] Difference between LIMMA and Genespring
>
> Dear Sir/Madam,
>
> The bioinformaticist whom we cooperated with analyzed the Agilent 
> microarray data with LIMMA and Genespring. As what she said, LIMMA 
> should be more stringent than Genespring, so that we can get more genes 
> with Genespring analysis, but in fact we get less genes by Genespring 
> sometimes and sometimes ¼ of the genes are different, even we both use 
> ANOVA and Benjamini & Hochberg for statistic analysis. So my question 
> is:

> 1 Is this kind of situation reasonable?

> 2 What' the reason for the different results between LIMMA and Genespring?
>
> Best wihses!
>
> Ming
>
> ________________________________
> Helmholtz Zentrum München
> Deutsches Forschungszentrum für Gesundheit und Umwelt (GmbH)
> Ingolstädter Landstr. 1
> 85764 Neuherberg
> www.helmholtz-muenchen.de
> Aufsichtsratsvorsitzende: MinDir´in Bärbel Brumme-Bothe
> Geschäftsführer: Prof. Dr. Günther Wess und Dr. Nikolaus Blum
> Registergericht: Amtsgericht München HRB 6466
> USt-IdNr: DE 129521671

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:5}}