To the Maintainers of "Agi4x44PreProcess" and the Bioconductor community We are using custom-designed Agilent chips (4 x 45220 probe) to analyse one-colour samples prepared from zebrafish embryos. The question I have concerns the use of annotation databases in "Agi4x44PreProcess". The raw text files generated by the Agilent Feature Extraction software contain the probe and gene name for each spot on the array. We also have a custom designed gal file. The problem we have is that only 5 zebrafish-specific annotation databases exist for zebrafish: four of these have been designed for use with the Affymetrix zebrafish chip, whilst the remaining database uses entrez ids. Our gene names are predominantly listed as ENSEMBL transcript identifiers (only a small proportion are entrez ids). Therefore there doesn't yet exist an appropriate annotation database for use with our data. With this in mind I have two questions: Firstly, is it possible to run the Agi4x44PreProcess software successfully without calling an annotation database (can the probe ids within the .txt files be used to identify the replicated probes)? Secondly, if it is not possible to run the Agi4x44PreProcess software without using an annotation database, is there any way that we could somehow use the gal file instead? I apologise in advance if these are naive questions - I am a newbie to microarrays and all thing r and would appreciate any advice that you can give me. Many thanks in anticipation of your help. Best Wishes Samantha (Sam) England Department of Physiology, Development & Neuroscience University of Cambridge

Hi Samantha, Have you considered just generating an annotation package using the SQLForge code in AnnotationDbi? There is a SQLForge vignette explaining how to do this here as well as other explanations on how to use the annotation packages: http://www.bioconductor.org/packages/release/bioc/html/AnnotationDbi.h tml For those ensembl transcript IDs, you could convert them to entrez gene annotations 1st using the org.Dr.egENSEMBLTRANS mapping in the org.Dr.eg.db package and then simply follow the example in the vignette above. Let us know if you have questions about how to do this stuff after looking over the AnnotationDbi vignettes. Marc On 07/16/2010 08:20 AM, Samantha England wrote: > The raw text files generated by the Agilent Feature Extraction software contain the probe and gene name for each spot on the array. We also have a custom designed gal file. The problem we have is that only 5 zebrafish-specific annotation databases exist for zebrafish: four of these have been designed for use with the Affymetrix zebrafish chip, whilst the remaining database uses entrez ids. Our gene names are predominantly listed as ENSEMBL transcript identifiers (only a small proportion are entrez ids). Therefore there doesn't yet exist an appropriate annotation database for use with our data. With this in mind I have two questions: