Hello,
I just recently used the SCAN.UPC package because I need to normalize some data from Affymetrix's Equine Gene 1.0 ST Arrays using custom CDF from Michigan's BrainArray, and this seemed the only package that I could use. Very nice package, pretty good documentation and easy to use! However, I couldn't find it explicitly stated anywhere in the vignette, the original paper, the SCAN help page nor searching on the support site what scale the resulting normalized values are on. From their distribution, I'm assuming it's a log scale, but since the range of values for my samples is only from -1.8 to 5.2, I assume it's not the standard log2? The equations in the paper use log() in several places, so it's on the natural log scale? Either way, please add the information to the documentation.
Thanks,
Jenny
A slightly orthogonal point; Benilton added functionality to the pdInfoBuilder package intended primarily to accommodate the MBNI re-mapped probeset definitions. There isn't yet a vignette for this, but I will probably add one in devel soon. In the meantime, a short version goes like this.
Download the 'Zip of CDF, Seq, Map, Desc' file from MBNI (The last column here, if you are using e.g., Entrez Gene mappings). I will illustrate with Equus, since that is what you are using. Unzip the file somewhere, then start R in that directory.
## setup library(pdInfoBuilder) library(foreach) library(doMC) ## if you have multiple cores registerDoMC(4) ## or more, depending on how cool you are ## make package unzip("equgene10st_EQca_ENTREZG_19.0.0.zip") ## this step uses multiple cores if you have them tab <- cdf2table("equgene10st_EQca_ENSG.cdf") ## probably you want to use reasonable email and author names, unlike me... pd <- new("GenericPDInfoPkgSeed", table = tab, pkgName = "pd.equgeneMBNI.1.0.st", version = "0.0.1", email = "me@mine.com", author = "me") makePdInfoPackage(pd) ## install and use install.packages("pd.equgeneMBNI.1.0.st/", repos = NULL) library(oligo) ## you have to use the pkgname argument here! dat <- read.celfiles(list.celfiles(), pkgname = "pd.equgeneMBNI.1.0.st") eset <- rma(dat)