[BioC] 2-way Anova analysis

[Naomi Altman]

Dear Sharon,
I think that what you mean is that you want to consider the treatment 
effect within strain.

To do this, you simply create 2 contrasts - one for each strain.  The 
coefficients for the other strain will be 0.

--Naomi

At 04:08 AM 1/6/2006, Sharon wrote:
>Hi Donglei & James
>
>
>Thanks for your reply.  I am sorry for not making myself clear. I already
>have this matrix that contains Probeset    p-value (strain effect) p-value
>(treatment effect) p-value (interaction).  As you suggested, I can use
>pvalue cut-off to select probesets for strain effect. From the selected list
>(for strain effect), I would like to choose probsets that are significant
>for Strain A and then for Strain B.  In the same way, I would like to select
>probesets for Treatment A and Treatment B.
>
>I hope, this explains my problem.
>Thanks again,
>Sharon
>
>
>On 1/5/06, donghu at itsa.ucsf.edu <donghu at itsa.ucsf.edu> wrote:
> >
> > Hi,
> >
> > You can make a matrix which has 4 columns:
> >
> > Probeset    p-value (strain effect) p-value (treatment effect) p-value
> > (interaction)
> >
> > Then you can select probe sets based on these p-values.
> >
> > Hope this helps.
> >
> > Donglei Hu
> >
> > On Thu, 5 Jan 2006 15:26:15 +0100 Sharon wrote:
> >
> > >  Hi All,
> > >
> > > I am trying to analyze affy data for strain, treatment and their
> > > interaction
> > > effects (RMA normalized, log2 expression values). The design looks like
> > >
> > >    strain A --> treatment 1       strain B --> treatment 1
> > >                     treatment 2                         treatment 2
> > >
> > > so, I used anova(aov(expra ~ strain + treatment + strain*treatment, m)),
> > > where m <- data.frame(strain, treatment, expr)
> > >
> > > I have got the corresponding pvalues for strain, treatment & interaction
> > > effects( this for all the probsets).  Now, I would like to separate the
> > > number of significant probesets for strain A and strain B from strain
> > > effect
> > > (I would like to do the same for treatment effect).  How can I do
> > > this? Any
> > > help is very much appreciated.
> > >
> > > Thanks in Advance.
> > > Regards,
> > >  Sharon
> > >
> > >       [[alternative HTML version deleted]]
> > >
> > > _______________________________________________
> > > Bioconductor mailing list
> > > Bioconductor at stat.math.ethz.ch
> > > https://stat.ethz.ch/mailman/listinfo/bioconductor
> > >
> >
> >
> >
>
>         [[alternative HTML version deleted]]
>
>_______________________________________________
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch
>https://stat.ethz.ch/mailman/listinfo/bioconductor

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