[BioC] MAS5 present/absent calls

Eric Blanc eblanc at ebi.ac.uk
Mon Mar 6 11:39:55 CET 2006


Hi,

The MAS5 original documentation 
(http://www.affymetrix.com/support/technical/whitepapers/sadd_whitepaper.pdf) 
  describes how the Present/Absent calls are computed. This document 
emphasizes that the cutoff values (alpha1=0.04 and alpha2=0.06) for 
attributing a P/M/A call given a P value are suitable for probesets 
which have between 16 and 20 probes. The man page for mas5calls states 
that the bioconductor implementation in the affy package uses the same 
values. I have 2 questions about that:

1) What are the guidelines when the number of probes per probeset is 
lower (11 to 14) ?
2) Does anybody knows what is the default MAS5 behaviour when the number 
of probes per probeset is lower ?

Thanks for your help,

Eric



More information about the Bioconductor mailing list