[BioC] Discrepancy on results from gcrma function and justGCRMA

Jin jzhang at vapop.ucsd.edu
Wed Oct 28 23:31:36 CET 2009


Jerry <norn2k at ...> writes:

> 
> Hi,
>  
> I'm currently using gcrma package in R to summarize probeset intensities from 
CEL files. Surprisingly,
> I found that the results generated from gcrma function and justGCRMA function 
are quite different. In
> general, expression values from gcrma function are lower and the boxplots from 
gcrma are quite
> asymmetric (no tails on the bottom). I attached some plots below for your 
information. This confused
> me as I thought that the two functions implemented similar algorithms. 
>  
> Thank you so much for your help!
>  
> Best,
> Jianjun
> 
> PS: The package I used is gcrma 2.12.1. I also observed similar results 
on 2.14 version.
> 
>       

Hello,

I'm running into the same problem with gcrma and justGCRMA. They are not giving 
the same numeric results. Is this an intended feature? 

I was under the impression that these two methods were identical with the 
exception that justGCRMA was more memory efficient because it didn't have to 
create an AffyBatch object.

Can anyone shed some light on this subject?

Thanks,
Jin



More information about the Bioconductor mailing list