[BioC] [LIMMA]lmFit with continuous predictor

[Belinda Phipson]

Hi Jack

I would fit my design matrix like this:

> design<-model.matrix(~A+B)

and because there are only two levels for A it is treated as a factor, but B
is treated as a continuous variable.

> design
  (Intercept) A  B
1           1 1 25
2           1 0 35
3           1 0 28
4           1 1 32

Then proceed as usual:

> fit<-lmFit(data,design)
> fit<-eBayes(fit)
> summary(decideTests(fit))

# For significant genes for effect A:
> topTable(fit,coef=2)

# For significant genes for effect B:
> topTable(fit,coef=3)

Cheers,
Belinda

-----Original Message-----
From: bioconductor-bounces at r-project.org
[mailto:bioconductor-bounces at r-project.org] On Behalf Of Yao Chen
Sent: Friday, 13 July 2012 7:08 AM
To: bioconductor at r-project.org
Subject: [BioC] [LIMMA]lmFit with continuous predictor

Hi all,

I am trying to use Limma for multiple regression. The design matrix is like
this :

              EffectA   EffectB
                  1            25
                  0            35
                  0            28
                  1            32

The Effect A is categorical, The Effect B is continuous.  I can use
contrast.matrix  to find the differential expressed genes between 0 and 1
(EffectA). But I don't know how can I find genes correlated with EffectB.

Thanks,

Jack

	[[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor at r-project.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives:
http://news.gmane.org/gmane.science.biology.informatics.conductor


______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}