- From: Vittorio Meloni <vittorio.meloni@crs4.it>
- Date: Wed, 7 Apr 2021 09:29:47 +0200
- To: jerven Bolleman <jerven.bolleman@sib.swiss>
- Cc: public-bioschemas@w3.org
- Message-ID: <CAFqAGsyzhOC6T0ehDb0mx_2n6w0=OV0VT7o=EAR2OXv3-9VbrA@mail.gmail.com>
Hi Jerven,
thank you very much for the feedback and suggestions.
The use of variableMeasured to express additional attributes not provided
by the model was actually one of the major doubts I had.
I've adopted it because it is the only one I've found for Dataset in
Bioschema profile.
I agree that additionalProperty seems to fit better but I've found it only
in the deprecated DataRecord type.
I'll consider also the "specialist" way.
Thanks again,
Vittorio
Il giorno mar 6 apr 2021 alle ore 15:39 jerven Bolleman
<jerven.bolleman@sib.swiss> ha scritto:
> Hi Vittorio,
>
> Take this as suggestions, as I do expect that some people would model
> things differently than I would here.
>
>
> You reuse a number of ontologies, where the url is the ID.
> e.g.
> {
> "@type": "CategoryCode",
> "codeValue": "OBI_0000922",
> "name": "frozen specimen",
> "url": "http://purl.obolibrary.org/obo/OBI_0000922",
> ->>>> "@id": "http://purl.obolibrary.org/obo/OBI_0000922"
> }
>
>
> You are inconsistent with ICD 10 references
>
> "value": "urn:miriam:icd:C34.9",
> in one place.
> "http://identifiers.org/icd/C34.9"
> in another
> should really be
> "https://id.who.int/icd/release/10/C34.9"
>
> Which is not free, but is the IRI of the thing in the official
> ICD repositories. In any case pick one and be consistent about it.
> If you want to use identifiers.org use it for everything, not
> identifiers.org for some things and obolibrary for other things.
> An option here is
>
> {
> "@id" : "https://id.who.int/icd/release/10/C34.9",
> "url" : "https://icd.who.int/browse10/2010/en#/C34.9"
> }
>
> My personal opinion here is to also stop with schema.org and start with
> domain specific ontologies. For example the schema:diagnosis property
> looks enticing but would be wrong to use. I also don't think the
> schema:variableMeasured is correct.
>
> The diagnosis is not a measured variable of the dataset but of the
> original patient around the time the tissue sample was taken. If I was
> staying within (bio)schema.org I would use schema:additionalProperty
> even if it is not preferred, because being specialist I don't think
> there is a good generic solution.
>
> To be honest adding a bit of context so that you can use specialist
> ontologies seems better to me. Something in the spirit of this "wrong"
> modeling.
>
>
> "@context": [ "https://schema.org",
> { "diagnosis" : { "@id":
> "http://purl.obolibrary.org/obo/OGMS_0000073"},
> "icd10":"https://id.who.int/icd/release/10/"
> }
> ],
> "@type": "Dataset",
> "@id": "bbmri-eric:ID:UK_GBR-1-56:collection:162-526",
> ->>>> "diagnosis": "icd10:C34.9",
> "name": "Neoplastic pathways",
>
> Diagnosis is a owl:class abused as a property here, which is not "good"
> but I unfortunately don't have time to look for an existing property
> which accurately expresses your needs.
>
>
> Regards and good luck,
>
> Jerven
>
>
> On 06/04/2021 14:33, Vittorio Meloni wrote:
> > Hi Jerven,
> > yes of course. You can find attached an example of a Biobank and a
> > Collection one.
> >
> > Thanks,
> > Vittorio
> >
> > Il giorno mar 6 apr 2021 alle ore 14:12 jerven Bolleman
> > <jerven.bolleman@sib.swiss> ha scritto:
> >
> > Hi Petr, Vittorio,
> >
> > Would you mind sharing some markup/json-ld so that we have a concrete
> > example to work with. For me at least a concrete example helps a lot.
> >
> > Regards,
> > Jerven
> >
> > On 06/04/2021 14:07, Petr Holub wrote:
> > > Dear all,
> > >
> > > could you please review Vittorio's proposal below? We are
> > implementing
> > > the BioSchemas into production in the BBMRI-ERIC Directory and the
> > > initial mapping we have done almost 3 years back are no longer
> > usable -
> > > and thus Vittorio has created the proposal below and we would
> > like to
> > > confirm it before continuing.
> > >
> > > Thanks,
> > > Petr
> > >
> > >
> > >
> > > **Assoc. Prof. *RNDr. Petr Holub, Ph.D.
> > > *Senior IT/Data Protection Manager*
> > >
> > > BBMRI-ERIC* | Neue Stiftingtalstrasse 2/B/6 | 8010 Graz |
> AUSTRIA
> > > Phone: +43 316 34 99 17-18 | Fax: +43 316 34 99 17-99 | Mobile:
> > +43 664
> > > 88 72 18 77
> > > Skype:holubp | petr.holub@bbmri-eric.eu
> > <mailto:petr.holub@bbmri-eric.eu>
> > > <mailto:petr.holub@bbmri-eric.eu
> > <mailto:petr.holub@bbmri-eric.eu>> | www.bbmri-eric.eu
> > <http://www.bbmri-eric.eu>
> > > <http://www.bbmri-eric.eu
> > <http://www.bbmri-eric.eu>><
> https://mail.acpcloud.de/owa/redir.aspx?C=tQtD-G9C_kuPP8hViFQ4hgNjuhtf_dEI0JIE-vC5gXhf36JcThZMI5N8J9ocveR1mgFita1orqo.&URL=http%3a%2f%2fwww.bbmri-eric.eu%2f
> > <
> https://mail.acpcloud.de/owa/redir.aspx?C=tQtD-G9C_kuPP8hViFQ4hgNjuhtf_dEI0JIE-vC5gXhf36JcThZMI5N8J9ocveR1mgFita1orqo.&URL=http%3a%2f%2fwww.bbmri-eric.eu%2f
> >>
> > >
> > >
> >
> ------------------------------------------------------------------------
> > > *From:* Vittorio Meloni <vittorio.meloni@crs4.it
> > <mailto:vittorio.meloni@crs4.it>>
> > > *Sent:* Tuesday, March 30, 2021 14:09
> > > *To:* public-bioschemas@w3.org <mailto:public-bioschemas@w3.org>
> > <public-bioschemas@w3.org <mailto:public-bioschemas@w3.org>>
> > > *Subject:* Advice for applying Bioschemas to BBMRI Directory
> > > Dear Bioschemas Members,
> > > I am Vittorio Meloni and I'm collaborating with BBMRI to add
> > Bioschemas
> > > into the Biobanks' Directory. In particular, I need to represent
> > > Biobanks and Collections concepts, which are compliant with the
> > MIABIS
> > > 2.0 model: the Biobank is the organization that stores samples
> > and data
> > > related to the samples while a Collection is the set of samples.
> > > According to these definitions, I mapped the Biobank to
> DataCatalog,
> > > using the "provider" attribute to describe the organization and
> the
> > > "dataset" attribute for the list of Collections, and the
> > Collection to a
> > > Dataset. Do you think this mapping would be correct?
> > > Another issue I have is how to add additional properties to a
> > > Collection: for example, how can I express that a Collection
> > collects
> > > materials as DNA? With the DataRecord I could use
> > additionalProperty,
> > > but it is Deprecated now. I've found the variableMeasured
> > property but
> > > I'm not sure it fits correctly. Do you think that's the right way
> > to go?
> > >
> > > Any suggestion or comment would be appreciated.
> > > Thank you,
> > > Vittorio
> > >
> > > --
> > > Vittorio Meloni
> > > Data Intensive Computing - Digital Health Group
> > > CRS4 - Centro di Ricerche, Sviluppo e Studi Superiori della
> Sardegna
> > > POLARIS, Edificio 1, 09010 PULA (CA - Italy)
> > >
> > > Email: vittorio.meloni@crs4.it <mailto:vittorio.meloni@crs4.it>
> > > <https://webmail.crs4.it/imp/message.php?index=221#
> > <https://webmail.crs4.it/imp/message.php?index=221#>>
> >
> > --
> >
> > *Jerven Tjalling Bolleman*
> > Principal Software Developer
> > *SIB | Swiss Institute of Bioinformatics*
> > 1, rue Michel Servet - CH 1211 Geneva 4 - Switzerland
> > t +41 22 379 58 85
> > Jerven.Bolleman@sib.swiss - www.sib.swiss
> >
> >
> >
> >
> > --
> > Vittorio Meloni
> > Data Intensive Computing - Digital Health Group
> > CRS4 - Centro di Ricerche, Sviluppo e Studi Superiori della Sardegna
> > POLARIS, Edificio 1, 09010 PULA (CA - Italy)
> >
> > Email: vittorio.meloni@crs4.it
> > <https://webmail.crs4.it/imp/message.php?index=221#>
>
> --
>
> *Jerven Tjalling Bolleman*
> Principal Software Developer
> *SIB | Swiss Institute of Bioinformatics*
> 1, rue Michel Servet - CH 1211 Geneva 4 - Switzerland
> t +41 22 379 58 85
> Jerven.Bolleman@sib.swiss - www.sib.swiss
>
>
>
--
Vittorio Meloni
Data Intensive Computing - Digital Health Group
CRS4 - Centro di Ricerche, Sviluppo e Studi Superiori della Sardegna
POLARIS, Edificio 1, 09010 PULA (CA - Italy)
Email: vittorio.meloni@crs4.it
<https://webmail.crs4.it/imp/message.php?index=221#>
Received on Wednesday, 7 April 2021 07:30:18 UTC