[BioC] Question regarding background correction in lumi package

Pan Du dupan at northwestern.edu
Mon Feb 14 17:47:17 CET 2011


Hi Jeanie

> Hi Dr. Pan,
> 
> I am trying to use lumi package on the microarray data.  My questions are:
> 
> Q1. If I understand it right, the Beadstudio output data has to be
> background corrected then read in with lumiR, right? Then what does
> 'bgAdjust' do in lumiB on an already background-corrected data? Was it
> supposed to be applied on a raw data (without background subtraction)?

Background correction is NOT required before importing to the lumi pipeline.
You can use background adjustment methods implemented in the lumiB function.
The 'bgAdjust' background adjustment method has similar implementation as
BeadStudio, which basically estimates the background level based on the
negative control probes.

> Q2. If I want to use log2 transformation, shall I specify method
> 'forcePositive' in lumiB or I just leave it as "none" to do nothing on the
> bacground-substracted data and assume lumiT step will take care of that
> automatically?

If you use log2 transformation, I recommend using forcePositive in lumiB
because NA will produced if the value is negative. Actually, I don't
recommend using background level adjustment if you use log2 transformation
because variation close to zero will be very high after log2 transformation.
Vst transformation will automatically deal with negative value problem.

Hope this explanation is clear to you.


Pan



> 
> Thanks for your help in advance.
> 
> Jeanie  
> 
> 
> ===================================
> 
> P Please consider the environment before printing this e-mail
> 
> Cleveland Clinic is ranked one of the top hospitals
> in America by U.S.News & World Report (2010).
> Visit us online at http://www.clevelandclinic.org for
> a complete listing of our services, staff and
> locations.
> 
>



More information about the Bioconductor mailing list