[BioC] lmFit: how method I use in non-Affy single channel

Gordon Smyth smyth at wehi.edu.au
Fri Apr 8 13:33:28 CEST 2005

>Date: Thu, 07 Apr 2005 15:20:02 -0300
>From: Marcelo Luiz de Laia <mlaia at fcav.unesp.br>
>Subject: [BioC] lmFit: how method I use in non-Affy single channel
>To: "'bioconductor at stat.math.ethz.ch'"
>         <bioconductor at stat.math.ethz.ch>
>Dear limma users,
>The function lmFit have 3 methods. One, default, use gls.series and
>another, robust, use rlm.series from MASS package. In help, gls.series
>is mentioned to be used in hight correlation between spots replication.
>In my case I dont have this.
>I visited the help files but I dont get knowledge for decide how I use
>in my data set. The method robust show more differentially genes and a
>"B" value around 50. The P.value was around 1E-016. This is  OK?
>Please, if you have a time, tell me how this method I use. Or I try they
>and use how I like?

Some people always like to use robust statistical methods when analysing 
microarray data, and the "robust" option to lmFit() is provided for this 
reason. There is no rule which tells you which data to use it for and which 


>I have a single channel data set from filters (nylon membrane). Each
>experiment have 6 biological repetition and each gene is printed 2 times
>in each filter. I have one reference and 3 times points. I construct a
>contrast matrix and do all constrast. I use topTable and heatDiagram to
>get genes.

More information about the Bioconductor mailing list