22

rq turtle/ttl

• gene ontology
• identifier

For two accession numbers (ACs) find the GO term labels and group them into GO process,function and component

Use at

• https://sparql.uniprot.org/sparql/

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX up: <http://purl.uniprot.org/core/>
PREFIX GO:<http://purl.obolibrary.org/obo/GO_>

SELECT 
    (CONCAT(SUBSTR(STR(?protein), 33)) AS ?uniprot)
    (GROUP_CONCAT(?celtype; separator=";") AS ?celtypes)
    (GROUP_CONCAT(?biotype; separator=";") AS ?biotypes)
    (GROUP_CONCAT(?moltype; separator=";") AS ?moltypes)
WHERE
{
    VALUES (?ac) {("Q6GZX4") ("Q96375")}
    BIND (IRI(CONCAT("http://purl.uniprot.org/uniprot/",?ac)) AS ?protein)
    ?protein a up:Protein .
    ?protein up:classifiedWith ?goTerm .
    #Determine if the type is biological_process
    OPTIONAL {
        ?goTerm rdfs:subClassOf GO:0008150 .
        ?goTerm rdfs:label ?biotype .
    }
    #Determine if the type is cellular_component
    OPTIONAL {
        ?goTerm rdfs:subClassOf GO:0005575 .
        ?goTerm rdfs:label ?celtype .
    }
    #Determine if the type is molecular_function
    OPTIONAL {
        ?goTerm rdfs:subClassOf GO:0003674 .
        ?goTerm rdfs:label ?moltype .
    }
    #Filter out the uniprot keywords
    FILTER(bound(?biotype) || bound(?celtype) || bound(?moltype))
} GROUP BY ?protein

graph TD
classDef projected fill:lightgreen;
classDef literal fill:orange;
classDef iri fill:yellow;
  v4("?ac")
  v1("?biotype"):::projected 
  v9("?biotypes")
  v2("?celtype"):::projected 
  v8("?celtypes")
  v6("?goTerm")
  v3("?moltype"):::projected 
  v10("?moltypes")
  v5("?protein"):::projected 
  v7("?uniprot")
  c5(["GO:0008150"]):::iri 
  c8(["GO:0003674"]):::iri 
  c7(["GO:0005575"]):::iri 
  c2(["up:Protein"]):::iri 
  f0[["(bound(?biotype) || (bound(?celtype) || bound(?moltype)))"]]
  f0 --> v1
  f0 --> v2
  f0 --> v3
  bind1[/VALUES ?ac/]
  bind1-->v4
  bind10(["Q6GZX4"])
  bind10 --> bind1
  bind11(["Q96375"])
  bind11 --> bind1
  bind2[/"concat('http://purl.uniprot.org/uniprot/',?ac)"/]
  v4 --o bind2
  bind2 --as--o v5
  v5 --"a"-->  c2
  v5 --"up:classifiedWith"-->  v6
  subgraph optional0["(optional)"]
  style optional0 fill:#bbf,stroke-dasharray: 5 5;
    v6 -."rdfs:subClassOf".->  c5
    v6 --"rdfs:label"-->  v1
  end
  subgraph optional1["(optional)"]
  style optional1 fill:#bbf,stroke-dasharray: 5 5;
    v6 -."rdfs:subClassOf".->  c7
    v6 --"rdfs:label"-->  v2
  end
  subgraph optional2["(optional)"]
  style optional2 fill:#bbf,stroke-dasharray: 5 5;
    v6 -."rdfs:subClassOf".->  c8
    v6 --"rdfs:label"-->  v3
  end
  bind6[/"concat(substring(str(?protein),'33^^xsd:integer'))"/]
  v5 --o bind6
  bind6 --as--o v7
  bind7[/"?celtype"/]
  v2 --o bind7
  bind7 --as--o v8
  bind8[/"?biotype"/]
  v1 --o bind8
  bind8 --as--o v9
  bind9[/"?moltype"/]
  v3 --o bind9
  bind9 --as--o v10