Property talk:P715

From Wikidata
Jump to navigation Jump to search

Documentation

DrugBank ID
identifier in the bioinformatics and cheminformatics database from the University of Alberta
Descriptionidentifier in the DrugBank (Q1122544) database
Applicable "stated in" valueDrugBank (Q1122544)
Data typeExternal identifier
Template parameterTemplate:Infobox drug (Q6033882): en:Template:Drugbox: DrugBank
Domain
According to this template: chemical substance (Q79529)
According to statements in the property:
type of chemical entity (Q113145171) or group of isomeric entities (Q15711994)
When possible, data should only be stored as statements
Allowed valuesDB(SALT\d|CAT\d)?\d{5}
Examplevitamin C (Q199678)DB00126
Sourcehttps://www.drugbank.ca
Formatter URLhttps://go.drugbank.com/drugs/$1
https://www.drugbank.ca/r/$1
Tracking: sameno label (Q32085200)
Tracking: differencesCategory:P715 different on Wikidata (Q20636191)
Tracking: usageCategory:Pages using Wikidata property P715 (Q26250011)
Tracking: local yes, WD noCategory:P715 absent from Wikidata (Q20636190)
Lists
Proposal discussionProposal discussion
Current uses
Total25,641
Main statement12,987 out of 13,491 (96% complete)50.6% of uses
Qualifier2<0.1% of uses
Reference12,65249.3% of uses
Search for values
[create Create a translatable help page (preferably in English) for this property to be included here]
Format “DB(SALT\d|CAT\d)?\d{5}: value must be formatted using this pattern (PCRE syntax). (Help)
Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/P715#Format, SPARQL
Single value: this property generally contains a single value. (Help)
List of violations of this constraint: Database reports/Constraint violations/P715#Single value, hourly updated report, SPARQL
Distinct values: this property likely contains a value that is different from all other items. (Help)
Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/P715#Unique value, SPARQL (every item), SPARQL (by value)
Scope is as main value (Q54828448), as reference (Q54828450): the property must be used by specified way only (Help)
List of violations of this constraint: Database reports/Constraint violations/P715#Scope, hourly updated report, SPARQL
Allowed entity types are Wikibase item (Q29934200): the property may only be used on a certain entity type (Help)
Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/P715#Entity types

Format

[edit]

Format of this property should be changed to include pages like [1]. Wostr (talk) 19:33, 28 September 2018 (UTC)[reply]

I agree. Drugbank has separate listings for salt forms. Currently the majority of the constraint violations (Wikidata:Database reports/Constraint violations/P715) occur because the items for the neutral form and for the salt form both link to the same Drugbank page even though there are separate pages for the two types. I don't see why the salts shouldn't be directly linked to the corresponding salt pages instead. The "Allowed values" would then be best changed to allow an identifier such as "SALT000272" for your example. Edgar181 (talk) 20:08, 15 December 2018 (UTC)[reply]
It would probably require to change all values to DBxxxxx. I don't understand why 'DB' prefix is not included, but for many other databases similar prefixes are included. Maybe {{Autofix}} could be used, but first we had to inform all the Wikimedia projects that use this property about the changes. Then we could change the format constraint (Q21502404) to match ids like DB/SALTxxxxx. Wostr (talk) 20:32, 15 December 2018 (UTC)[reply]
I'va changed the constraint to "(SALT\d{1})?\d{5}". Wostr (talk) 23:14, 11 January 2019 (UTC)[reply]

I think these changes are not sufficient. Please keep some things in mind:

  • We are not talking just about salts (altough the ID suggests). In fact, Drugbank calls entries like that "Product Ingredients" because there may also be different esters etc.
  • Each SALT-ID is related to an active compound with a DB-ID, in most cases for both of them a CAS number is also given. We should distinguish between the Drug Bank active compound record and several "Product Ingredient" records with different properties.
  • Each Product Ingredient is linked to an active compound, each active compound may have a couple of different product incredients. Links between these two types of DrugBank entries should also be present here.

What do you think?--Mabschaaf (talk) 14:25, 6 July 2019 (UTC)[reply]

As far as I know, there should not be any problems with DrugBank ids (either DB... or SALT/xxx...) in WD, because every pure compound, salt of this compound, product with this compound as an ingredient should have its own WD item. If there is the same CAS number in two DrugBank entries, it should be investigated what caused this situation. Constraint needs to be modified it there are any other ids we want to include in WD.
Also, 'DB' prefix should be added to every value in WD; I don't know why there is no 'DB' here and prefixes are present in e.g. Human Metabolome Database ID (P2057) which is very similar database. Lack of 'DB' prefixes also causes a problem with using this property outside WD, e.g. in Wikipedia's infoboxes, because we need to add this prefix manually in the code (but other re-users of WD data may not know that they need to add 'DB' prefix). Wostr (talk) 14:56, 6 July 2019 (UTC)[reply]
I agree with you on the "DB" prefix.
My question/suggestion was, to add to all items with SALT-IDs a statement like amlodipine mesylate (Q27254365) (should have DBSALT001964) "is an ingredient of" amlodipine (Q411347) (DB00381). Maybe there is already a qualifier matching this situation.--Mabschaaf (talk) 10:55, 7 July 2019 (UTC)[reply]
Statement amlodipine mesylate (Q27254365) "is an ingredient of" amlodipine (Q411347) is not quite correct — these are two different compounds, one is not an ingredient of another, but I think I know what you want to accomplish here. This is a much larger problem that was not sorted out during importing data about pharmaceutical products. Right now we have a pair active ingredient in (P3780)/has active ingredient (P3781) that should connect a pharmaceutical product with an active substance. We do not have dedicated property for linking between pharmaceutical form←→pharmaceutical product and pharmaceutical form←→active substance. But right now even active ingredient in (P3780)/has active ingredient (P3781) pair is not used as intended, because sometimes we have pharmaceutical form instead of active substance. On the example of morphine (Q81225) (active substance), morphine sulfate (Q27107392) (pharmaceutical form of an active substance) and Duramorph (Q47521513) (pharmaceutical product) we should be able to link between:
It can be simplified to:
So we lack a pair 'used in a pharmaceutical form as'/'is a pharmaceutical form of' and has part(s) (P527) statements in every pharmaceutical product. That is a huge problem that hasn't been addressed nor fixed during the data import. It could be done like this:
and everything about an active substance (morphine (Q81225)) could be easily queried. However, this is not a problem with DrugBank ID (P715) and I don't think we can solve it here. Wostr (talk) 13:28, 7 July 2019 (UTC)[reply]
Yes, I think, you got my point. I agree that this is not directly related to the question about the IDs. How can we proceed on that? We could either add the "DB" hopefully in an automated process, but we would create lots of IDs not matching the real situation. Maybe it would be best to reimport all IDs from the drugbank database itself, match them with the CAS numbers, overwrite the present entries and add also sources. I would really prefer that way.--Mabschaaf (talk) 18:14, 7 July 2019 (UTC)[reply]
@Wostr: Any progress on this? --Mabschaaf (talk) 18:05, 28 July 2019 (UTC)[reply]
I am currently importing all IDs freshly (with "DB"). With link to the matched InChIKey. If someone feels like removing the old ones and adapting the property, cool. Else I'll remove then gradually later AdrianoRutz (talk) 13:04, 23 May 2024 (UTC)[reply]
[edit]

Hi all, the format regular expression suggests DBCAT005101 is now a valid identifier. The formatter URL, however, is not suited for this. I just tweeted the DrugBank team to see if they have a URL pattern that we can use instead and supports regular DBxxx drug identifiers and the DBCATxxx category identifiers. --07:42, 30 May 2020 (UTC)