[BioC] gcrma

Lana Schaffer schaffer at scripps.edu
Thu Jan 6 18:48:28 CET 2005


Do you have any benchmarks for theses options?  Are they in your
published papers?
Thanks,
Lana Schaffer


>===== Original Message From Zhijin Wu <zwu at jhsph.edu> =====
>The "fullmodel" option uses both MM intensities and PM probe sequences;
>The  "affinities" option uses MM probes only as negative control probes to
>estimate the relationship of NSB with sequence. IT does not use the MMs as
>paried measurement of background for each PM probe. The idea here is that
>a set of "non-PM" probes are needed but not one for each PM.
>The "MM" option uses only the MM measurements for background adjustment.
>
>
>On Wed, 5 Jan 2005, Lana Schaffer wrote:
>
>> Hi,
>> I have been reading the papers "A Model Based Background Adjustment for
>> Oligonucleotide Expression Arrays" and "Stochastic Models Inspired by
>> Hybridization theory for short Oligonucleotide Arrays" and comparing
>> these papers to the options available in Bioconductor for gcrma.
>> I request some clarification for the options: fullmodel, affinities,
>> mm, and constant.  Does the fullmodel use the MLE or the Bayes estimate?  
Is
>> there some literature that compares the results of the various options to 
one
>> another, since the options don't seem to be exactly those presented in the
>> papers?  I assume the mm option to use an unadjusted mm?
>> So would you say that gcrma still uses the mm's?
>> Thanks for your input.
>> Lana
>>
>> Lana Schaffer
>> Research Specialist
>> The Scripps Research Institute
>> DNA Array Core Facility
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>

Lana Schaffer
Research Specialist
The Scripps Research Institute
DNA Array Core Facility



More information about the Bioconductor mailing list