From: Valerie W. <va...@sa...> - 2002-07-31 14:21:32
|
Angel Pizarro wrote: > > Arnaud, > this is a bit over powered for what we use, but I think it not all > that much more, except the bit about the authors. Can I make the > compromise to keep the author field in the table in addition to a > normalized set of tables for those that choose to use them? > > ALso I had forgotten to add an (external_db_release_id, source_id) tuple > to the table before I sent it out. This would only take care of one > external reference, though. A possible solution would be to use a many to > many relation to DbRef for all of the external ids such as pubmed and > medline. > > Lastly, what is your intent with the 'is_curated' bit column? > Angel We need to be able to add a flag to distinguish between papers which have been read and manually curated, and those which have been mined but not curator reviewed to extract data Val -- ---------------------------------------------------------------------------------- Valerie Wood Tel: 01223 494954 S. Pombe Genome Project Fax: 01223 494919 The Sanger Centre email: va...@sa... Wellcome Trust Genome Campus http://www.sanger.ac.uk/Projects/S_pombe Cambridge CB10 1SA |