[BioC] lumi question, and a bug

Michal Blazejczyk michal.blazejczyk at mail.McGill.ca
Fri Jun 22 17:46:24 CEST 2007


Hi,

I have a bug report and a question regarding the lumi package.

1. Question: are the annotations, target IDs etc. used in any
   of the calculations, or are they only provided for descriptive
   reasons and for downstream analysis?  I'm asking because in
   our workflow we don't read data files using lumiR, and we
   don't provide annotations - instead we construct LumiBatch
   objects from matrices of data.

2. Bug report: theoretically, the presence of bead counts and
   detection p-values in input data files should be optional.
   However, when one reads the code of function lumiR, it first
   checks whether they are present but later assumes that they
   are.  A LumiBatch object also cannot be constructed without
   these values - if I provide NULL instead I get:
     Error in function (storage.mode = c("lockedEnvironment", "environment",  :
     'AssayData' elements with invalid dimensions: 'beadNum' 'detection'
   I think that beadNum and detection should be made fully
   optional.

Best,
Michal



More information about the Bioconductor mailing list