Dear list, I am using an affymetrix porcine data set and, before differential expression using limma, I would like to use non-specific filtering using nsFilter function from genefilter package. I understood that there is no annotation package for porcine arrays (unless I am wrong), so, one choice is to build an annotation package to be used in the nsFilter function from the original Affymetrix annotation file. For this purpose, I think I can use the makeDBPackage function from AnnotationDbi package, but I am not able to manage it. I've tried with the following chunk of code: library(pig.db0) makeDBPackage("PIG_DB", affy=TRUE, prefix="Porcine", fileName="Porcine.na30.annot.csv", baseMapType="gb", outputDir = "C:/", version="1.0.0", manufacturer = "Affymetrix", chipName = "Porcine", manufacturerUrl = "http://www.affymetrix.com") The csv file is the one related to porcine annotation downloaded from Affymetrix and it is in the same working directory. The result of executing this code is NULL. I've tried also with the following code makeDBPackage("HUMANCHIP_DB", affy=FALSE, prefix="hcg110", fileName=hcg110_IDs, baseMapType="gb", outputDir = tmpout, version="1.0.0", manufacturer = "Affymetrix", chipName = "Human Cancer G110 Array", manufacturerUrl = "http://www.affymetrix.com") and it works, so, clearly, it is something wrong in my code that I'm not able to see... Any suggestions? Thanks in advance, Javier

Hi Javier, You are VERY close. The main trouble is that your schema is the wrong kind. You want to use a chip schema (PIGCHIP_DB) because you are making a chip package. The other kind of schema (PIG_DB, HUMAN_DB etc.) is reserved for making org packages, and we already make those for you. Here is what I used (note that the outputDir is for unix): makeDBPackage("PIGCHIP_DB", affy=TRUE, prefix="Porcine", fileName="Porcine.na30.annot.csv", baseMapType="gbNRef", outputDir = ".", version="1.0.0", manufacturer = "Affymetrix", chipName = "Porcine", manufacturerUrl = "http://www.affymetrix.com") This should create a source directory for you called Porcine.db Once you have the source directory, you should be able to build, check and install it as with any other source package for R. Marc On 03/24/2010 03:38 PM, Javier P?rez Florido wrote: > Dear list, > I am using an affymetrix porcine data set and, before differential > expression using limma, I would like to use non-specific filtering > using nsFilter function from genefilter package. > I understood that there is no annotation package for porcine arrays > (unless I am wrong), so, one choice is to build an annotation package > to be used in the nsFilter function from the original Affymetrix > annotation file. > > For this purpose, I think I can use the makeDBPackage function from > AnnotationDbi package, but I am not able to manage it. I've tried with > the following chunk of code: > > library(pig.db0) > > makeDBPackage("PIG_DB", > affy=TRUE, > prefix="Porcine", > fileName="Porcine.na30.annot.csv", > baseMapType="gb", > outputDir = "C:/", > version="1.0.0", > manufacturer = "Affymetrix", > chipName = "Porcine", > manufacturerUrl = "http://www.affymetrix.com") > > The csv file is the one related to porcine annotation downloaded from > Affymetrix and it is in the same working directory. The result of > executing this code is NULL. I've tried also with the following code > > makeDBPackage("HUMANCHIP_DB", > affy=FALSE, > prefix="hcg110", > fileName=hcg110_IDs, > baseMapType="gb", > outputDir = tmpout, > version="1.0.0", > manufacturer = "Affymetrix", > chipName = "Human Cancer G110 Array", > manufacturerUrl = "http://www.affymetrix.com") > > and it works, so, clearly, it is something wrong in my code that I'm > not able to see... > > Any suggestions? > Thanks in advance, > Javier > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor >