[BioC] Simple minfi question

[Gustavo Fernández Bayón]

Hi Kasper.  

Thank you for your answer. It is more than enough. I wanted just to be sure about what I was doing.

Regards,
Gus


---------------------------
Enviado con Sparrow (http://www.sparrowmailapp.com/?sig)


El miércoles 19 de septiembre de 2012 a las 15:50, Kasper Daniel Hansen escribió:

> If you have two groups, doing an F test is equivalent to doing a
> t-test. If you have more than two groups and want to test if they are
> all the same, you need to do an F-test. You can think of the F-test
> as a generalization of the t-test. A more detailed answer is surely
> out of the scope of this email list.
>  
> Kasper
>  
> On Wed, Sep 19, 2012 at 5:23 AM, Gustavo Fernández Bayón
> <gbayon at gmail.com (mailto:gbayon at gmail.com)> wrote:
> > Hi everybody.
> >  
> > I have been playing a little bit with the minfi package, using some 450k data we have received. I have seen that dmpFinder function uses a F-test when looking for differentially methylated positions (when type is 'categorical').
> >  
> > Until now, as a newbie, I have been usually searching for these DMP's with t-tests and posterior multiple testing corrections. I have been looking for info but, as always, my lack of statistics knowledge is becoming a major handicap.
> >  
> > Question is, why F-test instead of t-test? Are both methods valid? Or is there a specific reason for this preference?
> >  
> > By the way, my congratulations to Kasper Daniel Hansen and Martin Aryee for their amazing work. I have several projects coming with 450k data, and minfi will surely be of great help.
> >  
> > Regards,
> > Gustavo
> >  
> >  
> > ---------------------------
> > Enviado con Sparrow (http://www.sparrowmailapp.com/?sig)
> >  
> > _______________________________________________
> > Bioconductor mailing list
> > Bioconductor at r-project.org (mailto:Bioconductor at r-project.org)
> > https://stat.ethz.ch/mailman/listinfo/bioconductor
> > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>