[MOBY-l] I3C Life sciences identifier

Brian Gilman gilmanb at genome.wi.mit.edu
Fri May 17 19:12:36 UTC 2002 

On 5/17/02 1:46 PM, "Mark Wilkinson" <mwilkinson at gene.pbi.nrc.ca> wrote:
Hey Guys,

    Mark is correct, I'm the co-chair of the technical architecture working
group of the I3C (TechArch). This group is committed to creating and
adopting standards that *actually work*. This has been a personal gripe of
mine when trying to use other standards.

    I am going to wear my I3C hat now and try to answer some of your
questions and address your gripes. We understand that the I3C has not
produced anything substantial in its brief life. This was mostly because of
legal issues that have mired the I3C's path and efficacy. However, these
issues have been taken care of and we are truly ready to move forward at top
speed. In fact, we are putting together a demo which uses the LSID
specification as a backbone. I am also working very hard to get this demo
and the hardware and software that is part of it into the communities hands
to try out. My deliverables are:

            A SOAP service to do comparative genomics:

             1) NCBI --> BSML XSLT
             2) Java/Perl API's to produce and consume BSML
             3) Software to parse and produce LSID's
             4) API's that speak SOAP to the I3C service demo

    The I3C is not a closed body of venders!! in fact, many academic
institutions were invited to the initial meetings. The WICGR and NCI were
very early participants in the i3c discussions as well as early contributors
to some of the standards that are being put forth for community discussion
and comment. (LSID/BSML/AGAVE etc.)

    To this end, I would like to personally invite the MOBY group to submit
its specifications to the I3C for review and potential adoption. Without
support from groups like MOBY/DAS/BioJava/BioPerl/GO/HL7 etc. The I3C would
be missing a very large, technically gifted and seasoned group of domain
experts! 

                        -B
-- 
Brian Gilman <gilmanb at genome.wi.mit.edu>
Group Leader Medical & Population Genetics Dept.
MIT/Whitehead Inst. Center for Genome Research
One Kendall Square, Bldg. 300 / Cambridge, MA 02139-1561 USA
phone +1 617  252 1069 / fax +1 617 252 1902



    
    
> Hi Michael,
> 
> Brian Gilman, a BioMOBY project member and developer, is on the Technical
> Development (committee?) for I3C.  It is through him that we have an inkling
> of
> what the I3C is doing.
> 
> <personal_gripe>
> The fact that it *is* difficult to discover what the I3C is doing (and your
> obvious discomfort in revealing their plans speaks to this) seems to also make
> the community as a whole uncomfortable in my experience...  Given that they
> (we)
> are the ones who will be asked put these unknown standards into practice, I
> can
> understand that there is some objection to not being involved in the creation
> process... but that is just my personal opinion;  I am not speaking for anyone
> else involved in the BioMOBY project.  Certainly, this is the main reason that
> all BioMOBY development is being done in the open with as much community
> involvement as possible.
> </personal_gripe>
> 
> 
> Iny any case, Brian gave us an overview of the LSID at the last MOBY meeting.
> We
> considered it, played with it, and revamped our ideas somewhat to accomodate
> some
> of the attributes of the LSID, but we didn't adopt it whole-sale.  As we've
> said
> on this mailing list several times, none of us want to ignore other standards,
> and it may be that we migrate closer to, or adopt wholeheartedly, the I3C's
> LSID
> in the future, particularly if it ever does become an accepted standard.  As
> of
> this moment, however, we have a similar identifier in the MOBY scheme that
> seems
> sufficient to do the job at hand.
> 
> Having said that, perhaps this is a good time to revisit this issue.  We
> discounted the LSID fairly quickly at the last meeting in favour of the
> MOBY-Triple, and moving the AuthURI (and potentially other attributes of a
> message) into the MOBY-Envelope.  Does anyone think that we should reconsider
> this before we go too far?
> 
> Cheers all!
> 
> Mark
> 
> 
> Michael Gribskov wrote:
> 
>> I have received a draft spe for a "life sciences identifier" from the
>> Interoperable Informatics Infrastructure Consortium.    This seems like an
>> effort similar to moby and likely to be relevant.  the proposal specifically
>> mentions that it addresses a problem with the DAS standard not handling
>> "Query
>> by Identifier".  I guess my questions are:
>> 
>> Is this group in contact with moby?
>> Does this proposal work/not work with the moby idea of a service?
>> 
>> The gist of the proposal is as follows (for the whole text I think I should
>> refer you to the authors; I am uncomfortable posting the whole thing)
>> 
>> The Life Sciences Identifier (LSID) specification defines a scheme for
>> location-independent naming of distributed biologically significant data
>> items
>> such as a sequence, a clone, a gene, etc., as well as a mechanism by which
>> the
>> identity associated with the names can be authenticated without strictly
>> requiring a trusted third party. A biologically significant data item is
>> defined as a named data element (a file, a database row, etc.) that operates
>> in a distributed, federated environment and has a defined identity. LSID's
>> are
>> not meant to identify artifacts of implementation, like a database instance,
>> or a server, which should be handled by other mechanisms such as JDBC URLs or
>> WSDL.
>> LSID names are a type of Location-independent Object Identifier (LOID) that
>> uniquely identifies a biologically significant data item. The actual location
>> (address) of a data item is determined by associating the LOID with one or
>> more communication protocols and network endpoints. Collectively the set of
>> communication protocols and network endpoints for a data item is called its
>> ServiceDescription.
>> 
>> URN:LOID:LSID:<authority>:<namespace>:<objectID>:<securityInfo>
>> 
>> where
>> <authority>
>> The name of the organization that has defined the entity.
>> <namespace>
>> One or more statements that constrain the scope in which this
>>                  id is evaluated.
>> <objectID>
>> An alphanumeric sequence that uniquely defines this object
>>                   within this namespace, as defined by the authority.
>> <securityInfo>
>> An optional field containing a cryptographic identity.
>> --
>> Michael Gribskov
>> San Diego Supercomputer Center & Dept. of Biology
>> University of California, San Diego
>> 9500 Gilman Drive
>> La Jolla  CA  92093-0537
>> Tel:858.534.8312   Fax:858.822.0873
>> Email:gribskov at sdsc.edu    http://www.sdsc.edu/~gribskov
>> 
>> _______________________________________________
>> moby-l mailing list
>> moby-l at biomoby.org
>> http://biomoby.org/mailman/listinfo/moby-l
> 
> --
> --------------------------------
> "Speed is subsittute fo accurancy."
> ________________________________
> 
> Dr. Mark Wilkinson
> Bioinformatics Group
> National Research Council of Canada
> Plant Biotechnology Institute
> 110 Gymnasium Place
> Saskatoon, SK
> Canada
> 
> 
> 
> _______________________________________________
> moby-l mailing list
> moby-l at biomoby.org
> http://biomoby.org/mailman/listinfo/moby-l
>

More information about the moby-l mailing list