Dear Bioc users, This list has regularly announcements of new packages and discussion on fRMA and related approaches, so I thought this might be of interest. How to preprocess 20,000 CEL files on an ordinary desktop computer in a few hours? We recently published a new, fully scalable preprocessing method for(Affymetrix and other) short oligonucleotide microarray atlases (NAR 2013) <http: nar.oxfordjournals.org="" content="" 41="" 10="" e110="">. The method scales up to arbitrarily large collections of tens of thousands (or more) CEL files. The scalability is based on sequential, probabilistic hyperparameter updates, thus circumventing the extensive memory requirements of standard approaches. The method outperforms the standard RMA in various tests, and it is readily applicable also for the large majority of platforms where pre-calculated fRMA<http: biostatistics.oxfordjournals.org="" content="" 11="" 2="" 242.abstract="">parameter sets are not available, with a comparable performance. Online-RPA is freely available as a R/Bioconductor package<http: bioconductor.org="" packages="" release="" bioc="" html="" rpa.html="">. The wiki site <https: github.com="" antagomir="" rpa="" wiki=""> provides installation instructions and usage examples. For feedback, issues, bug tracking, and pull requests, see the Github development version<https: github.com="" antagomir="" rpa=""> . You can also have a look at the brief blog post<http: antagomir.wordpress.com="" 2013="" 10="" 20="" rpa-fully-scalable-="" preprocessing-method-for-short-oligonucleotide-microarray-atlases=""/> . All feedback concerning the package is mostly welcome. best regards, Leo Lahti, University of Helsinki, Finland [[alternative HTML version deleted]]