Wikidata queries

Examples of Wikidata queries relevant to stuff I work on. Includes discussion of how to add data, and federated queries.

Source(s) of a taxon name in Wikidata

Data sources for a taxonomic name

PREFIX prov: <http://www.w3.org/ns/prov#>
PREFIX pr: <http://www.wikidata.org/prop/reference/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>

SELECT *
WHERE
{
  ?wikidata wdt:P225 "Mammalia" .
  ?wikidata p:P225 ?statement. 
  ?statement prov:wasDerivedFrom ?provenance .
  ?provenance pr:P248 ?reference .
  ?reference wdt:P1476 ?title .
}

Try it

Find author of taxon name

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX pq: <http://www.wikidata.org/prop/qualifier/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX p: <http://www.wikidata.org/prop/>

SELECT *
WHERE
{
  ?wikidata wdt:P225 "Zodariidae" .
  ?wikidata p:P225 ?statement .
  ?statement pq:P405 ?authority .
  ?authority rdfs:label ?name .
  
  FILTER (lang(?name) = 'en')
}

Try it

Find taxa named by author (inverse of previous query)

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX pq: <http://www.wikidata.org/prop/qualifier/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX p: <http://www.wikidata.org/prop/>

SELECT *
WHERE
{
  ?statement pq:P405 wd:Q1346792 .
  ?taxon p:P225 ?statement .
  ?taxon wdt:P225 ?taxon_name .  
}

Try it

In this example wd:Q1346792 is Tamerlan Thorell.

Find date of publication

Date of publication of name

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX pq: <http://www.wikidata.org/prop/qualifier/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX p: <http://www.wikidata.org/prop/>

SELECT *
WHERE
{
  ?wikidata wdt:P225 "Liphistiidae" .
  ?wikidata p:P225 ?statement .
  ?statement pq:P574 ?date .
}

Try it

Find work based on DOI

SELECT *
WHERE
{
  ?work wdt:P356 "10.2476/ASJAA.62.33" .
}

Try it

If DOI not found can add it using http://tools.wmflabs.org/sourcemd

Find work based on DOI (case insensitive) SLOW

Wikidata has DOIs in UPPERCASE, CrossRef recommends lowercase, can use filter to have case insensitive query (slow)

SELECT *
WHERE
{
  ?work wdt:P356 ?doi .
  FILTER (lcase(str(?doi)) = "10.2476/asjaa.62.33")
}

Try it

Find journal from ISSN

SELECT *
WHERE
{
  ?journal wdt:P236 "0074-0276" .
}

Try it

Find Wikidata item for Wikispecies author

Wikispecies authors are on Wikidata, so we can get Wikidata items using the Wikispecies URL. with author name URL encoded (instead of underscores "_" use url encoded spaces %20).

SELECT *
WHERE
{
    VALUES ?article {<https://species.wikimedia.org/wiki/Bing_Zhou>}
	  ?article schema:about ?author .
    ?author wdt:P31 wd:Q5 .
}

[Try it] (http://tinyurl.com/yazywu3y)

Find an author based on name

SELECT ?_human ?name  WHERE { 
  VALUES ?family {"Agassiz"@en }
  VALUES ?given {"Alexander"@en }

  ?_human wdt:P31 wd:Q5.
  ?_human rdfs:label ?name .
                
  ?_human wdt:P734 ?familyName .
  ?familyName rdfs:label ?family . 
                
  ?_human wdt:P735 ?givenName .
  ?givenName rdfs:label ?given . 
  
   FILTER (lang(?name) = 'en')
}

Try it

Find an author based on name and birth and death dates

SELECT ?_human ?name  WHERE { 
  
  VALUES ?family {"Agassiz"@en }
  VALUES ?given {"Alexander"@en }
  VALUES ?birth { 1835 }
  VALUES ?death { 1910 }
  
  ?_human wdt:P31 wd:Q5.
  ?_human rdfs:label ?name .
                
  ?_human wdt:P734 ?familyName .
  ?familyName rdfs:label ?family . 
                
  ?_human wdt:P735 ?givenName .
  ?givenName rdfs:label ?given . 
  
  ?_human wdt:P569 ?birth_date .
  ?_human wdt:P570 ?death_date .
  
  FILTER(year(?birth_date) = ?birth) .
  FILTER(year(?death_date) = ?death) .
  FILTER (lang(?name) = 'en')

}

Try it

Generate list and map of museums and their collection codes

Based on the new Biodiversity Repository ID property we could list all museums and herbaria with codes and put them on a map. Only two examples at time of writing:

SELECT *  WHERE { 
  ?museum wdt:P4090 ?code .
  ?museum rdfs:label ?name .
  OPTIONAL {
  ?museum wdt:P625 ?coord .
  }
                
  FILTER (lang(?name) = 'en')
}

Try it

Adding data

To add a publication based on its DOI go to http://tools.wmflabs.org/sourcemd and paste in the DOI. If the DOI does not exist in Wikidata, you should redirected to the QuickStatements tool and see something like this:

CREATE
LAST	P356	"10.1016/J.SAJB.2012.07.014"
LAST	P31	Q13442814
LAST	P1476	en:"Species limits in Vachellia (Acacia) karroo (Mimosoideae: Leguminoseae): Evidence from automated ISSR DNA “fingerprinting”"
LAST	Len	"Species limits in Vachellia (Acacia) karroo (Mimosoideae: Leguminoseae): Evidence from automated ISSR DNA “fingerprinting”"
LAST	P304	"36-43"
LAST	P478	"83"
LAST	P577	+2012-11-00T00:00:00Z/10
LAST	P2093	"C.L. Taylor"	P1545	"1"
LAST	P2093	"N.P. Barker"	P1545	"2"

These are triples describing the article. “LAST” links every triple to the Wikidata identifier created for the article.

Below you will see the text:

You need to authorize WiDaR to edit Wikidata on you behalf for this tool to work!

https://tools.wmflabs.org/widar/index.php?action=authorize

Authorise WiDaR and then add the statments to Wikidata (click on "Do it"). You should see the results of your editing below:

Processing Q32062483 (Q32062483 P356 "10.1016/J.SAJB.2012.07.014")
Processing Q32062483 (Q32062483 P31 Q13442814)
Processing Q32062483 (Q32062483 P1476 en:"Species limits in Vachellia (Acacia) karroo (Mimosoideae: Leguminoseae): Evidence from automated ISSR DNA “fingerprinting”")
Processing Q32062483 (Q32062483 Len "Species limits in Vachellia (Acacia) karroo (Mimosoideae: Leguminoseae): Evidence from automated ISSR DNA “fingerprinting”")
Processing Q32062483 (Q32062483 P304 "36-43")
Processing Q32062483 (Q32062483 P478 "83")
Processing Q32062483 (Q32062483 P577 +2012-11-00T00:00:00Z/10)
Processing Q32062483 (Q32062483 P2093 "C.L. Taylor" P1545 "1")
Processing Q32062483 (Q32062483 P2093 "N.P. Barker" P1545 "2")
All done!.

Remove statements

This can be done using QuickStatements 2 see https://www.wikidata.org/wiki/Help:QuickStatements#Removing_statements

Add repository code (Institution code) to Wikidata

Go to QuickStatements, make sure you have authorised WiDar, and enter statements. This example adds "NHMUK" to the entry for the Natural History Museum, London.

Q309388	P4090	"NHMUK"

List articles published in a journal

In this example we list all the articles published in Zootaxa (ISSN 1175-5326).

SELECT * WHERE {
  ?journal wdt:P236 "1175-5326".
  ?work wdt:P1433 ?journal.
  ?work wdt:P1476 ?title.
  ?work wdt:P478 ?volume.
  ?work wdt:P577 ?date .
  OPTIONAL { ?work wdt:P356 ?doi. }
} ORDER BY (year(?date))

Try it

Federated SPARQL

This query calls two different SPARQL endpoints, Uniprot and Wikidata. I gather Wikidata limits the possible external SPARQL series it can call, so this query needs to be run on another SPARQL endpoint. This query works on Apache Jena Fuseki Version 3.4.0 (running as a container stain/jena-fuseki ).

PREFIX up:<http://purl.uniprot.org/core/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>	
SELECT *
WHERE
{
  # Find taxon in NCBI taxonomy called "Begonia"
  # Convert IRI to string
  SERVICE <http://sparql.uniprot.org/sparql> 
  {
    ?taxon up:scientificName "Begonia" .
    BIND( REPLACE( STR(?taxon),"http://purl.uniprot.org/taxonomy/","" ) AS ?ncbi). 
  }
  
  # Find Wikidata entity for same taxon using NCBI tax_id
  SERVICE <https://query.wikidata.org/sparql> 
  {
    ?wikidata wdt:P685 ?ncbi .
    ?wikidata wdt:P225 ?name .
  }	  
}	

Result:

taxon	ncbi	wikidata	name
http://purl.uniprot.org/taxonomy/3681	"3681"	wd:Q158617	"Begonia"

For more examples and background see Integrating Wikidata and other linked data sources – Federated SPARQL queries

Can also run slightly modified version on the Uniprot SPARQL endpoint:

PREFIX up:<http://purl.uniprot.org/core/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>	
SELECT *
WHERE
{
  # Find taxon in NCBI taxonomy called "Begonia"
  # Convert IRI to string
  
  ?taxon up:scientificName "Begonia" .
  BIND( REPLACE( STR(?taxon),"http://purl.uniprot.org/taxonomy/","" ) AS ?ncbi). 

  # Find Wikidata entity for same taxon using NCBI tax_id
  SERVICE <https://query.wikidata.org/sparql> 
  {
    ?wikidata wdt:P685 ?ncbi .
    ?wikidata wdt:P225 ?name .
  }	  
}	

Try it

OpenURL-style queries

Find article from ISSN, volume, and first page

SELECT * WHERE 
{ 
  VALUES ?issn {"1175-5326" } .
  VALUES ?volume {"2528" } .
  VALUES ?firstpage {"^1(–[0-9]+)?$" } .
  
  ?work wdt:P1433 ?container .
  ?container wdt:P236 ?issn.
  ?work wdt:P478 ?volume .
  ?work wdt:P304 ?pages .
  FILTER regex(?pages,?firstpage,'i')
}

Try it

Note that ?firstpage is a regular expression to match (in this case) an article that starts on page 1.

Find article from journal, volume, first page

SELECT * WHERE {
  VALUES ?container_title { "Zootaxa"@en } .
  VALUES ?volume {"2528" } .
  VALUES ?firstpage {"^1(–[0-9]+)?$" } .
 
  ?work wdt:P1433 ?container.
  ?container rdfs:label ?container_title .
  ?work wdt:P478 ?volume.
  ?work wdt:P304 ?pages.
  FILTER regex(?pages,?firstpage,'i')
}

Try it

Note the need for the correct journal name and the language qualifier (@en).

Find by volume, first page, year

Can try a tuple of [volume, page, year] to avoid problems of string matching on journal name.

SELECT * WHERE {
  VALUES ?volume {"2528" } .
  VALUES ?firstpage {"^1(–[0-9]+)?$" } .
  VALUES ?year { "2010" } .
 
  ?work wdt:P1433 ?container.
  ?work wdt:P478 ?volume.
  ?work wdt:P304 ?pages.
  ?work wdt:P577 ?date .
  FILTER regex(?pages,?firstpage,'i') .
  FILTER(STR(YEAR(?date)) = ?year)
}

Try it

Find articles by author from their ORCID

Given an ORCID we can retrieve a list of articles a person has published.

SELECT * 
WHERE { 
  ?author wdt:P496 "0000-0002-5721-1840" .
  ?work wdt:P50 ?author .
  OPTIONAL {
    ?work wdt:P356 ?doi 
  }
  ?work wdt:P1476 ?title .
}

Try it

Get list of author name strings in order

Note the use of “p”, “ps”, and “pq” to get statment details and qualifier (in this case serial order).

SELECT ?work ?author_name_string ?author_order  WHERE 
{
  ?work wdt:P356 "10.1206/0003-0090(2006)297[0001:TATOL]2.0.CO;2" .
  OPTIONAL {
    ?work p:P2093 ?statement.
    ?statement ps:P2093 ?author_name_string .
    ?statement pq:P1545 ?author_order. 
  }
} ORDER BY (xsd:integer(?author_order))

try it

Find articles by author’s Wikidata id

SELECT * 
WHERE { 
  ?work wdt:P50 wd:Q22110755 .
  OPTIONAL {
    ?work wdt:P356 ?doi 
  }
  ?work wdt:P1476 ?title .
}

Try it

Find articles by author from Wikispecies id

SELECT *
WHERE
{
    VALUES ?article {<https://species.wikimedia.org/wiki/Erik_J._van_Nieukerken>}
	?article schema:about ?author .
    ?author wdt:P31 wd:Q5 .
    OPTIONAL {
      ?work wdt:P50 ?author .
      OPTIONAL {
        ?work wdt:P356 ?doi 
      }
      ?work wdt:P1476 ?title .
    }
}

Try it

Matching authors in Wikidata and Wikispecies

Author listed by string (P2093) only so get names to compare

# If successful this query will return the Wikidata item for an author 
# together with the author's name as recorded in Wikipsecies and
# in the Wikidata item for a work with a given DOI where the author is
# given as a string (P2093) not an item (P50). A client could use this
# to get the two names and compare them, if they match then we could construct a
# series of quick statements to update the Wikidata item for the work with the
# Wikidata item for the author.
SELECT ?wikidata_author ?wikidata_author_name ?author_name_string  
WHERE {
    # On the Wikipsecies page "Alice_Wells" she is listed as the second author 
    # for the work with DOI "10.11646/ZOOTAXA.3964.2.2"
	VALUES ?wikispecies_page {<https://species.wikimedia.org/wiki/Alice_Wells>} .
	VALUES ?wikispecies_order { "2" } .
	VALUES ?doi { "10.11646/ZOOTAXA.3964.2.2" } .

    # Find Wikidata item for Wikispecies author
	?wikispecies_page schema:about ?wikidata_author .
	?wikidata_author rdfs:label ?wikidata_author_name .

    # Find the Wikidata item for work with the DOI
	?work wdt:P356 ?doi .
  
    # Get name of author in same position as Wikispecies record
	?work p:P2093 ?statement.
	?statement ps:P2093 ?author_name_string .
	?statement pq:P1545 ?wikispecies_order. 

    # Restrict outselves to English language
	FILTER (lang(?wikidata_author_name) = 'en')   
} 

Try it

Author has Wikidata item linked to work by P50

# Query to return Wikidata item (if one exists) for author at a given position
# in the list of authors of a work
SELECT ?wikidata_author ?wikidata_author_name
WHERE {
    # DOI for work
	VALUES ?doi { "10.3897/ZOOKEYS.507.9536" } .
    # Position in list of authors
	VALUES ?wikispecies_order { "1" } .

    # Get work for DOI
	?work wdt:P356 ?doi .
	
    # Get details on author so client can check whethe rname matches
	?work wdt:P50 ?wikidata_author .
    ?wikidata_author rdfs:label ?wikidata_author_name .
	?work p:P50 ?statement .
	?statement pq:P1545 ?wikispecies_order . 

    # Restrict ourselves to English
    FILTER (lang(?wikidata_author_name) = 'en') 
}

Try it

Find works by author by name string

# Find works by author by name string
SELECT *
WHERE {
	VALUES ?author_name_string { "Simon Vitecek" } .
  
    ?statement ps:P2093 ?author_name_string .
    ?work p:P2093 ?statement.

 	?work wdt:P356 ?doi .
    OPTIONAL {
        ?work wdt:P356 ?doi 
    }
    ?work wdt:P1476 ?title . 
} 

Try it

Taxonomic tree

Query that fetches all edges in a tree rooted on Liphistiidae, based on query in d3sparql.js.

PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>
SELECT ?root_name ?parent_name ?child_name WHERE
{
 VALUES ?root_name {"Liphistiidae"}
 ?root wdt:P225 ?root_name .
 ?child wdt:P171+ ?root .
 ?child wdt:P171 ?parent .
 ?child wdt:P225 ?child_name .
 ?parent wdt:P225 ?parent_name .
}

Try it

IUCN Status

SELECT * WHERE 
{ 
  ?wikidata wdt:P141 ?status. 
  ?wikidata rdfs:label ?name .
  ?wikidata wdt:P225 ?taxon_name .
  ?status rdfs:label ?status_label .
  FILTER (lang(?name) = 'en') .
  FILTER (lang(?status_label) = 'en')
}
LIMIT 10

Works describing new species

Works having “main subject” “species novae”

SELECT *
WHERE
{
  ?work wdt:P921 wd:Q27652812 . 
  ?work wdt:P356 ?doi .
}

Try it

Type localities

SELECT *
WHERE
{
   ?taxon wdt:P31 wd:Q16521 . 
   ?taxon wdt:P225 ?taxon_name . 
   ?taxon wdt:P5304 ?type_locality .
   ?type_locality p:P625 ?statement.
   ?type_locality rdfs:label ?type_locality_name .
   ?statement ps:P625 ?geo .
   FILTER (lang(?type_locality_name) = 'en')
 }

Try it

Taxonomic hierarchy (tricky because not a tree)

GROUP_CONCAT

Makes path, but rode dis jumbled (no natural way to order the taxa)

SELECT ?taxon  (GROUP_CONCAT(?parentTaxonLabel;SEPARATOR="|") AS ?path) WHERE 
{
  
?taxon wdt:P225 'Arabidopsis thaliana' . 
  
# path, but not ordered in any sensible way
?taxon wdt:P171+ ?parentTaxon . 
?parentTaxon rdfs:label ?parentTaxonLabel 
  
FILTER(LANG(?parentTaxonLabel) = "en")
  
}
GROUP BY ?taxon 

Try it

Specify ranks wanted

Use path queries

SELECT ?taxon ?taxonLabel ?path ?genusLabel ?familyLabel WHERE 
{
  
?taxon wdt:P225 'Arabidopsis thaliana' . 
?taxon rdfs:label ?taxonLabel .
  
# genus
?taxon wdt:P171* ?genus . 
?genus wdt:P105 wd:Q34740 .
?genus rdfs:label ?genusLabel .
  
# family
?genus wdt:P171* ?family . 
?family wdt:P105 wd:Q35409 .
?family rdfs:label ?familyLabel .
        
# order
?family wdt:P171* ?order . 
?order wdt:P105 wd:Q36602 .
?order rdfs:label ?orderLabel .

# path
 BIND(CONCAT(?orderLabel, " / ", ?familyLabel, " / ", ?genusLabel ) AS ?path) .
  
FILTER(LANG(?taxonLabel) = "en")
FILTER(LANG(?genusLabel) = "en")
FILTER(LANG(?familyLabel) = "en")
FILTER(LANG(?orderLabel) = "en")
}

Try it

Other topics

Mountains in African Rift Valley

SELECT * WHERE {
  ?item wdt:P31 wd:Q8502.
  ?item rdfs:label ?name .
  ?item wdt:P625 ?geo.
  ?item wdt:P361 wd:Q81591.
  OPTIONAL { ?item wdt:P1886 ?id } .
  OPTIONAL { ?item wdt:P2044 ?height } .
  OPTIONAL { ?item wdt:P18 ?image } .
  OPTIONAL { ?item wdt:P2348 ?erupted . } .
  FILTER (lang(?name) = 'en')
}