Wikidata:Property proposal/produced by

produced by edit

   Done: produced by (P2849) (Talk and documentation)
DescriptionLinks a biologic/biochemical entity with the entity it has been produced by
Data typeItem
Domainbiochemically active items, e.g. cells
Examplemelanosome (Q1065756)melanocyte (Q247101)
Sourcescientific papers, databases, ontologies, stating a certain connection. E.g. ExoCarta
Robot and gadget jobsProteinBoxBot will use this for certain types of relations
Motivation

Although Wikidata has several generic properties, there is no property which allows to link a biological/biochemical product to its producer. For genes, this has been modeled quite well already (encodes (P688) and encoded by (P702)). But for all other things, no property seems to exists which could establish this link. Use cases are anything that a cell/organ/tissue/glad exkretes, be it exosomes, protein complexes (e.g. colesterol particles) milk, urine. This concept might seem broad, but as the biological details are often still lacking, therefore I think this would be a very valuable property for representing biological knowledge . Sebotic (talk) 09:49, 2 October 2015 (UTC)[reply]

Discussion

  Support --Andrew Su (talk) 15:21, 2 October 2015 (UTC)[reply]

  Support At the molecular biology level, it would be good to have some sort of relationship property for relationships further downstream of the gene and this property is a logical next step following the relationship properties linking genes/gene products. Many biosynthetic entities are products of multiple genes/gene products, and are not sufficiently represented by the currently existing/proposed properties. Eg- would be able to link Dopamine and the Dopamine Biosynthetic process. This property would offer more specificity than Natural Product of taxon (which is great for non-recombinant organisms). Gtsulab (talk) 15:44, 2 October 2015 (UTC)[reply]

  Support lv_ra (talk) 16:05, 2 October 2015 (UTC)[reply]

  Comment I'd like add an inverse "produces" property, which would have as a domain "processes" (biological or not). Another interesting related pair would be "process uses"/"used by process" to link a process to its input materials.   Question Do we agree on generalizing the domain ? author  TomT0m / talk page 16:21, 2 October 2015 (UTC)[reply]

I'm hesitant about creating this rather generic property for this purpose. There is another property already called "product" that seems suited to the general concept. I think my thought here is to create one that more specifically describes the nature of the biology you intend to capture. --I9606 (talk) 18:22, 2 October 2015 (UTC)[reply]
@I9606: Is there actually some practical reason not to use the generic property ? If the process item is a subclass of biological process, or a type of biological entity, then we know the products of this process captures the process nature of the process, is'nt it ? We could more simply expand the scope of the other property. (Although I think we should differentiate processes like parthenogenesis (Q183236) and biological entities). By the way we should find a qualifier to link the process by which an entity produces something, like
⟨ angiosperms (Q25314)      ⟩ produce Search ⟨ Q25314 ⟩
by process Search ⟨ Q183236 ⟩
to complete the model. author  TomT0m / talk page 08:15, 3 October 2015 (UTC)[reply]
@TomT0m: As the inverse property for 'produced by', 'product or material produced or service provided (P1056)' can be used. I initiated a discussion on its talk page. But as the inverse property of product or material produced or service provided (P1056) is manufacturer (P176), manufacturer sounds a bit awkward to me in the context of biology (for cells, tissues, etc.). If it could be renamed to 'producer' or 'producer' could be added as a alias and the scope could be extended, property 'produced by' would not be required. Why I think we need a property like this: For many products, it is still unclear what's the detailed process/source they are produced by and for many circumstances, simply being able to state thing X produced by Y can be very useful. Furthermore, the property biological process (P682) is rather reserved for Gene Ontology defined biological processes (see target items in its property definition). Sebotic (talk) 07:24, 4 October 2015 (UTC)[reply]
@I9606, Sebotic: OK, I get the issue : in one case we refer to an "instance" entity, in the other to a class of entities, per Help:Classification. Hence I'm in favor to use generic property anyway as opposed to specific domain properties, but that make clear that in one case the producer is a real world entity, like "Iphone : <produced by entity : Apple>" vs. Iphone : <fruit> : <type of producer : Plant> for example. author  TomT0m / talk page 10:33, 4 October 2015 (UTC)[reply]
@TomT0m, Sebotic: I support the use of the generic produces/product properties as long as the definitions of the originating properties can be expanded. I also like the optional extension of the qualifier to describe how the production happens. I do think there will be room for more specific properties, but happy to get things started with the more generic version and either utilize the qualifier pattern or sub properties to specialize as needed. --I9606 (talk) 20:24, 5 October 2015 (UTC)[reply]
@I9606: I agree, becoming more specific as new knowledge becomes available will be a good and realistic way to go Sebotic (talk) 20:37, 5 October 2015 (UTC)[reply]
@I9606, Sebotic, TomT0m:Be wary of combining two concepts because there is an English word that covers both. The same may not be true in other languages.Ford is a manufacturer (P176) of cars. Apples are natural product of taxon (P1582) apple trees. Proteins are encoded by (P702) genes. All of these are similar relationships. I think there is a case for a specialist property for biological processes so I   Support but I would like the property to have a name which makes it's specialist nature obvious. Unfortunately I can't think of one. The inverse property could be called 'Biological product of' but 'Biologically produced by' seems awkward, at least in English. Anyone else have any ideas? Joe Filceolaire (talk) 00:06, 4 November 2015 (UTC)[reply]
@Filceolaire: Which two concepts ? In one case, the product is the result of a transformation process, in the other the prouct is the result of the transformation process. It's the same concept and this has little to do with language peculiarities. author  TomT0m / talk page 12:34, 4 November 2015 (UTC)[reply]

  WikiProject Molecular biology has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead.

  Notified participants of WikiProject Medicine

  Oppose as is. “melanosome <produced by> melanocyte” doesn't make any sense. Melanosomes are an integral part of melanocytes, the relation should be “exists in” or “part of”. “secreted by” may be a better wording for the property. —Tinm (d) 01:53, 7 February 2016 (UTC)[reply]

@Tinm: Melanosomes are produced by Melanocytes and then exported as extracellular vesicles to keratinocytes [1], so your argument cannot be regarded as valid. Production of extracellular vesicles is a common process in biology, this is one reason why we need that property. Sebotic (talk) 18:03, 17 May 2016 (UTC)[reply]

@Sebotic, Andrew Su, Gtsulab, Emitraka, Tinm, TomT0m:  Done Majority support. Hopefully the description limits the use enough. As alias "secreted by" is included as suggested by Tinm. Lymantria (talk) 20:54, 22 May 2016 (UTC)[reply]