Scott, This is a fabulous idea, and one that I have been thinking over every time I read an academic publication. There is so much redundancy in most papers and this just contributes to the information overload problem. I agree with the notion of "asserting facts" and "providing provenance" about those facts. This becomes a sort of ad-hoc voting mechanism. Coupled with a small "journal ontology" where you can roughly say what section of a traditional journal the fact might fit into (introduction, discussion, findings, etc).
Do you know of a standardized provenance/depreciation ontology? We found the PROVENANCE VOCABULARY CORE ONTOLOGY here http://trdf.sourceforge.net/provenance/ns.html but aren't sure if there is a more robust and standard (widely used) ontology. -----Original Message----- From: public-semweb-lifesci-requ...@w3.org [mailto:public-semweb-lifesci-requ...@w3.org] On Behalf Of M. Scott Marshall Sent: Friday, June 24, 2011 6:13 AM To: HCLS Cc: r.a....@reading.ac.uk Subject: Nanopublications [excerpted from another discussion list, been meaning to bring this to your attention] The nice thing about nanopublications, see DOI link below, is that they lead very nicely to RDF. I would describe them as an assertion with prescribed provenance. If we want to share assertions about SNPs, then we need to be able to programmatically consume and assess them. In order to do so, they should use common vocabularies and meet provenance requirements. The handiest way to do this is with RDF and SPARQL. Of course, each different type of assertion, has special provenance requirements in order for consumers to decide if it meets their criteria. The SNP assertion (nanopublication) should refer to a reference sequence, a population/ethnicity, etc. A differentially expressed gene assertion (from an expression study) should refer to the experimental conditions, as well as a p or q-value, etc. As each of the different assertion types gets an RDF rendering, we can nanopublish it as linked data. And a new era of publishing is born. It is very encouraging to see what appears to be a mass movement to SPARQL endpoints taking place. At least, to my eyes, the movement appears relatively massive :) . NCBO has had a SPARQL endpoint since the beginning of last year. ArrayExpress is working towards it, with initial production of RDF for the Gene Expression Atlas. Uniprot is preparing one and it sounds like others such as EMBL are interested as well. I am also aware that the European Innovative Medicine Initiative (IMI) called OpenPhacts has started a demo project by getting participants to create SPARQL endpoints to their online resources. BTW, OpenPhacts was put together by the same Barend Mons as the one who instigated the nanopublications mentioned in the Nature Genetics article and the Concept Web Alliance. Cheers, Scott -- M. Scott Marshall, W3C HCLS IG co-chair, http://www.w3.org/blog/hcls http://staff.science.uva.nl/~marshall On Fri, Jun 24, 2011 at 12:43 PM, Rutger Vos <r.a....@reading.ac.uk> wrote: > On a slightly different note, but not irrelevant for us scientists & > semantic web aficionados, everyone seen this? It's about > micro-attribution to incentivize semantic annotation: > > Crowdsourcing human mutations > Nature Genetics, Vol. 43, No. 4. (29 April 2011), pp. 279-279 > http://dx.doi.org/10.1038/ng0411-279 > > -- > Dr. Rutger A. Vos > School of Biological Sciences > Philip Lyle Building, Level 4 > University of Reading > Reading, RG6 6BX, United Kingdom > Tel: +44 (0) 118 378 7535 > http://rutgervos.blogspot.com
