Jim -- well, that code executed just fine, so now I'm moving forward with using the resulting object and seeing if it works for me. Thanks so much! I didn't expect just one line of code to solve my problem. Jessica On 5/16/14 12:58 PM, James W. MacDonald wrote: > Hi Jessica, > > You have to build your own transcript package. Probably the easiest > thing to do is > > library(GenomicFeatures) > txdb.canFam3 <- makeTranscriptDbFromUCSC("canFam3","refGene") > > Best, > > Jim > > > On 5/16/2014 12:31 PM, Jessica Perry Hekman wrote: >> Hi, all. I am new to Bioconductor. I've tried hard to find the answer >> to my question in online documentation, so I apologize if I'm asking >> something which is already out there. >> >> I'm hoping to perform RNA-seq pathway analysis using GAGE and Pahview. >> The RNAseq workflow documentation here: >> >> http://bioconductor.org/packages/release/bioc/vignettes/gage/inst/doc /RNA-seqWorkflow.pdf >> >> >> shows that a known-genes annotation package is required for this >> workflow. The example uses human, TxDb.Hsapiens.UCSC.hg19.knownGene. >> Some other species have pre-built packages, but not dog. >> >> I'm having trouble figuring out how to construct the equivalent >> package for Canis familiaris 3.1. This vignette >> >> http://www.bioconductor.org/packages/release/bioc/vignettes/Genomic Features/inst/doc/GenomicFeatures.pdf >> >> >> (section 3) suggests that I can load a sqlite file based on UCSC known >> gene data. I can't quite figure out how to construct this sqlite file. >> Alternatively, I could load an annotation package such as >> TxDb.Hsapiens.UCSC.hg19.knownGene, and I'd be willing to build one of >> those for dog, but am not sure how. >> >> Any advice or pointers to appropriate documentation would be hugely >> appreciated. >> >> Thanks, >> Jessica >> >> > -- Jessica P. Hekman, DVM, MS PhD student, University of Illinois, Urbana-Champaign Animal Sciences / Genetics, Genomics, and Bioinformatics