Thanks Marc, this is good to know. We need to have the hg38 objects ready for general use -- there is nothing custom about using the same build as NHGRI. I wouldn't say there's a huge rush but there is a need for both a solution and a policy that addresses the management of an environment where analyses will be proceeding on two distinct builds. We straddled hg18 and hg19 without much difficulty but this was before specific addresses were so central to analysis. Things are different now and the core will have to devise a strategy for management, distribution, and real time selection of build in use -- IMHO.
On Mon, Jul 28, 2014 at 1:57 PM, Marc Carlson <mcarl...@fhcrc.org> wrote: > Hi Vince, > > The current Homo.sapiens package/object contains a reference to an hg19 > TranscriptDb object (which makes it specific to hg19). But you can easily > make a custom Homo.sapiens object of your own that points to a custom > TranscriptDb object. Making a custom OrganismDbi object is easy to do and > is documented in the OrganismDbi vignette: > > http://www.bioconductor.org/packages/devel/bioc/vignettes/ > OrganismDbi/inst/doc/OrganismDbi.pdf > > So there nothing very fancy going on here right now. But you can > customize to match a build if you have made the base annotation objects > (TranscriptDb, OrgDb etc.). > > > Marc > > > > > On 07/28/2014 10:37 AM, Vincent Carey wrote: > >> I confirmed that the rtracklayer liftOver will work with the appropriate >> chain. the >> remap utility seems to have some more configurability. might be worth a >> look. >> >> What I need to do promptly is put the proper seqinfo on result of >> makeCurrentGwascat >> so that folks are not led astray. We also need good marking of which >> build >> is in use on all our ranges, IMHO. >> >> >> On Mon, Jul 28, 2014 at 1:14 PM, Vincent Carey < >> st...@channing.harvard.edu> >> wrote: >> >> I just noticed that the addresses for NHGRI GWAS catalog distribution >>> are relative to GRCh38. Is there a plan for using Homo.sapiens with >>> selection of build? It may not be uncommon for someone to want to >>> work with different builds. >>> >>> What would be useful for me at this time is a simple way to get a Seqinfo >>> structure respecting GRCh38. Also, liftOver seems to have been replaced >>> by NCBI remap. Apparently the new build can also be called hg38, and a >>> chain >>> file exists, so rtracklayer liftOver utility should still succeed. >>> >>> [[alternative HTML version deleted]] >> >> _______________________________________________ >> Bioc-devel@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/bioc-devel >> > > _______________________________________________ > Bioc-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/bioc-devel > [[alternative HTML version deleted]] _______________________________________________ Bioc-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel
