Question: the URI for the "core" LinkedSPL graph

[rkboyce]

We have a question about how to create URIs for components of the LinkedSPLs resource. The "core" of the resources would represent all SPL sections and have predicates for active moieties and other items. Some of these will point to graphs that we develop as part of the project. We think that that main graph URI should be perhaps: http://linkedSPLs.bio2rdf.org#

And the other graph URIs like: http://linkedSPLs.bio2rdf.org/activeMoiety# http://linkedSPLs.bio2rdf.org/adverseEvents# etc.

Does this seem in line with current practice? We could also see using 'http://linkedSPLs.bio2rdf.org/core#' for the 'core' data so that all parts are named as distinct sub-graphs.

[micheldumontier]

Hi Rich! You can find our current guidelines for Bio2RDF URIs here: https://docs.google.com/document/d/1qXq27w_RER9Yjb9Z4WW8JmT09wWKGWX_Hbuhri4xQ9g/edit

in general, we use slash uris. our namespaces follow the form http://bio2rdf.org/[prefix]:[identifier] http://bio2rdf.org/[prefix]_vocabulary:[identifier] // for dataset-specific types and relations http://bio2rdf.org/[prefix]_resource:[identifier] // for dataset-specific resources not identified by source

for graphs, we current just use a single graph name that follows the form: http://bio2rdf.org/[prefix]_resource:bio2rdf-[prefix]-v[X]

[ningyifan]

Hi Michel,

I am a programmer work for Dr.boyce on the project bio2rdf/linkedSPLs.

Should we name the URI like belows?

For graphs: http://bio2rdf.org/linkedSPLs:xxx-xxx-xxxx http://bio2rdf.org/activeMoiety:xxxxx http://bio2rdf.org/adverseEvents:xxxxxxxxx http://bio2rdf.org/pharmacogenomics:xxx-xxx-xxxx http://bio2rdf.org/pDDI:xxxxxxxxxx

For relations: PREFIX linkedSPLs_vocabulary: http://bio2rdf.org/linkedSPLs_vocabulary:xxxx

For reources that don't have identifier: PREFIX linkedSPLs_resource: http://bio2rdf.org/linkedSPLs_resource:xxxx

Appreciate for any comments!

[micheldumontier]

Hi Yifan, my preference is that you keep all of the URIs in a single namespace (note: singularized & lowercase) http://bio2rdf.org/linkedspl:XXXX

if the identifiers overlap in their lexical space, then add prefixes e.g. am for active moiety, ae for adverse event, etc.

another option, is to use subnamespaces e.g. http://bio2rdf.org/linkedspl.activemoiety:xxxx

but then we need to register each one in our life science registry [1] and keep track that they are part of the same project

[1] tinyurl.com/lsregistry

m.

On Fri, Sep 5, 2014 at 11:53 AM, Yifan Ning notifications@github.com wrote:

Hi Michel,

I am a programmer work for Dr.boyce on the project bio2rdf/linkedSPLs.

Should we name the URI like belows?

For graphs: http://bio2rdf.org/linkedSPLs:xxx-xxx-xxxx http://bio2rdf.org/activeMoiety:xxxxx http://bio2rdf.org/adverseEvents:xxxxxxxxx http://bio2rdf.org/pharmacogenomics:xxx-xxx-xxxx http://bio2rdf.org/pDDI:xxxxxxxxxx

For relations: PREFIX linkedSPLs_vocabulary: http://bio2rdf.org/linkedSPLs_vocabulary:xxxx

For reources that don't have identifier: PREFIX linkedSPLs_resource: http://bio2rdf.org/linkedSPLs_resource:xxxx

Appreciate for any comments!

— Reply to this email directly or view it on GitHub https://github.com/bio2rdf/bio2rdf-scripts/issues/397#issuecomment-54666225 .

[ningyifan]

Got it. Thanks for your explanations. I will implements these URI changes ASAP.