From: Y. T. G. <yg...@pc...> - 2004-08-26 17:10:53
|
Hi Jinal, Your plan sounds reasonable to me. As for the question regarding RnaInstance, it is parallel to GeneInstance. You just need to make a RnaInstanceCategory entry (we have 0: unknown, 1: mRNA, 2: assembly <of ESTs>), make RnaInstance entries to be associated with this category and with RNAs. Also make RnaFeature using info such as rna name, number of exons to be attached to your RnaInstance. -Thomas > -----Original Message----- > From: gus...@li... > [mailto:gus...@li...] On Behalf > Of Jinal Jhaveri > Sent: Thursday, August 26, 2004 11:12 AM > To: gus...@li... > Cc: yg...@pc... > Subject: Re: [Gusdev-gusdev] dots.nagene to dots.gene > > > > Thanks Thomas, > > You reply was really helpful. I was thinking on the same > track. Here is what I > am planning to do. I already have one of my genomes annotated > (and submited > to genbank. And I was thinking of generating gbk files for > the other genomes > and submit it to gus using gbparser). So following are the steps I am > planning to take > > 1) Parse the genbank file and thus I will have most of my na > tables filled > 2) Fill the geneinstance and the gene tables and connect > geneinstance to > nafeatureimp using na_feature_id > 3) Fill the aa_featureimp table and connect it to nafeatureimp using > na_feature_id > 4) Fill the RNA table(just ids) and connecting it to gene table > 5) Fill the protein_instance table connect it to aafeatureimp through > aa_feature_id table > 6) Fill the protein table and establish relation with RNA > using rna_id and > ofcourse with protein_instance > > One thing which isn't clear is how will I fill in the > rnainstance table. How > can this information be available from ncbi genbank? Well I > understand that I > can ignore it for a while and just generate fake rna_ids just > to establish > connection between gene_ids and protein_ids (though it would > be suboptimal) > > Any suggestions/comments? > > > > > > > On Thursday 26 August 2004 11:40 am, Y. Thomas Gan wrote: > > > -----Original Message----- > > > From: gus...@li... > > > [mailto:gus...@li...] On Behalf Of > > > Steve Fischer > > > Sent: Wednesday, August 25, 2004 5:18 PM > > > To: Jinal Jhaveri > > > Cc: gus...@li... > > > Subject: Re: [Gusdev-gusdev] dots.nagene to dots.gene > > > > > > > > > To my knowledge we haven't done this yet. We are just > now starting > > > the planning phases of how to completely use the central dogma to > > > captures gene, geneinstances, rnas, etc. So far we have > only been > > > using it partially. > > > > > > Others have opinions? > > > > I think it is correct in that GUS does not yet have any plugin that > > populate the central dogma gene tables, and that annotators > make use > > of some of these tables. > > > > However in our Allgenes project (see allgenes.org), we have > pipelines > > to build DoTS transcripts (DTs) from ESTs, and DoTS Genes > (DGs) from > > DTs by aligning them to the genome. In the latter pipeline, I have > > plugins (not part of GUS yet) that does the following: > > 1) make a DoTS.GeneInstanceCategory entry (I made the very first > > non-placeholder entry, so I am pretty sure we have not used > this table > > much before) > > 2) make DoTS.GeneInstance entries for each of the DGs created by my > > pipeline, and give them the gene_instance_category_id I got above > > 3) associate each DoTS.GeneInstance with a corresponding > DoTS.Gene entry > > (for us, we already have a prior set of DoTS.Gene entries made by an > > orthogonal algorithm > > so I only had to map my gene instances to them instead of creating > > DoTS.Gene entries anew) > > > > I also created full sets of DoTS.GeneFeature, DoTS.RnaFeature, > > DoTS.ExonFeature, but I will ommit the details here since > it is pretty > > clear to see from the wiki page how to do this. > > > > -Thomas > > > > > steve > > > > > > On Aug 25, 2004, at 5:47 PM, Jinal Jhaveri wrote: > > > > Hi, > > > > > > > > As per my understanding the "gbparser" loads the uniq genes > > > > obtained from the genbank file in dots.nagene table. > Our group is > > > > planning to do annotation > > > > (and reannotation) of several genomes and thus we might > > > > > > have multiple > > > > > > > instances of the same gene and the reviewers might add > > > > > > comments etc. > > > > > > > All > > > > these required information is not available with > > > > > > dots.nagene but with > > > > > > > dots.gene (which make sense). I do understand the handling > > > > > > of central > > > > > > > dogma > > > > (thanks to the excellent explanation on WIKI) but what I don't > > > > understand is how to handle the transition between each > phase. How > > > > shall one load the genes > > > > obtained from the genbank file (i.e in dots.nagene) in > dots.gene (& > > > > dots.geninstance). Is it the matter of writing our own > > > > > > script/plugin > > > > > > > (I am > > > > fine with it), or am I missing a very basic step of the > > > > > > pipeline? I am > > > > > > > asking > > > > this because, in our project we are at a stage where I have to > > > > make decision (which might be very inefficient/wrong from the > > > > design > > > > > > perspective but > > > > > > > FAST > > > > as far as time is concerned) ,whether to change the GUS > > > > > > Schema and let > > > > > > > the > > > > data be in dots.nafeatureimp and dots.nagene table OR > to use the > > > > dots.gene, dots.geneinstance, etc tables and proceed logically. > > > > > > > > Please Help!!!!!! > > > > > > > > > > > > On the other note, I remember Steve mentioning about plans > > > > > > to arrange > > > > > > > a GUS > > > > Users Meeting in PCBI, in one of the mails. IMHO It would > > > > > > be great if > > > > > > > we can > > > > have that as I know atleast of 3 to 4 more groups (at Virginia > > > > Bioinformatics > > > > Institute) are planning to use GUS and everyone is waiting > > > > > > for a GUS > > > > > > > User > > > > Meeting like this where we can clear our doubts and become more > > > > efficient to handle it for our projects. I am sure other groups > > > > too will be interested in > > > > joining this. Any thoughts/comments on the plan ? > > > > > > > > > > > > > > > > ------------------------------------------------------- > > > > SF.Net email is sponsored by Shop4tech.com-Lowest price on > > > > > > Blank Media > > > > > > > 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for > > > > > > only $33 Save > > > > > > > 50% off Retail on Ink & Toner - Free Shipping and Free Gift. > > > > http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 > > > > _______________________________________________ > > > > Gusdev-gusdev mailing list Gus...@li... > > > > https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev > > > > > > ------------------------------------------------------- > > > SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank > > > Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic > > > DVD+R for only $33 Save 50% off Retail on Ink & Toner - Free > > > Shipping and Free Gift. > > > http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 > > > > > > _______________________________________________ > > > Gusdev-gusdev mailing list Gus...@li... > > > https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev > > > > ------------------------------------------------------- > > SF.Net email is sponsored by Shop4tech.com-Lowest price on > Blank Media > > 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for > only $33 Save > > 50% off Retail on Ink & Toner - Free Shipping and Free Gift. > > http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 > > _______________________________________________ > > Gusdev-gusdev mailing list Gus...@li... > > https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev > > > ------------------------------------------------------- > SF.Net email is sponsored by Shop4tech.com-Lowest price on > Blank Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic > DVD+R for only $33 Save 50% off Retail on Ink & Toner - Free > Shipping and Free Gift. > http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 > > _______________________________________________ > Gusdev-gusdev mailing list > Gus...@li... > https://lists.sourceforge.net/lists/listinfo/gusdev-gusdev > |