| // CC-BY 4.0 International. (c) 2024 Egon Willighagen | |
| @Grab(group='io.github.egonw.bacting', module='managers-ui', version='0.5.2') | |
| @Grab(group='io.github.egonw.bacting', module='managers-rdf', version='0.5.2') | |
| @Grab(group='org.openscience.cdk', module='cdk-smiles', version='2.9') | |
| @Grab(group='org.openscience.cdk', module='cdk-silent', version='2.9') | |
| @Grab(group='org.openscience.cdk', module='cdk-ctab', version='2.9') | |
| @Grab(group='org.openscience.cdk', module='cdk-sdg', version='2.9') | |
| import org.openscience.cdk.smiles.SmilesParser; |
| prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> | |
| prefix np: <http://www.nanopub.org/nschema#> | |
| prefix npa: <http://purl.org/nanopub/admin/> | |
| prefix npx: <http://purl.org/nanopub/x/> | |
| prefix xsd: <http://www.w3.org/2001/XMLSchema#> | |
| prefix dct: <http://purl.org/dc/terms/> | |
| select ?np ?label ?subj ?citationrel ?obj ?date where { | |
| graph npa:graph { | |
| ?np npa:hasValidSignatureForPublicKey ?pubkey . |
| #defaultView:Map | |
| SELECT ?osc ?oscLabel ?location WHERE { | |
| ?osc wdt:P1344 wd:Q104785223 . | |
| { | |
| ?osc wdt:P625 ?location . | |
| } UNION { | |
| ?osc wdt:P159 / wdt:P625 ?location . | |
| } | |
| SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". } | |
| } |
| new File("/tmp/foo.tsv").eachLine { line -> | |
| fields = line.split('\t') | |
| nanomileid = fields[0] | |
| ermid = fields[1] | |
| ermidShort = ermid.split(':')[1] | |
| content = """--- | |
| title: "Material: ${ermidShort}" | |
| type: ChemicalSubstance | |
| id: ${ermidShort} | |
| tag: ${ermid} |
| @Grab(group='org.bridgedb', module='org.bridgedb.bio', version='3.1.1') | |
| import java.util.ArrayList | |
| import java.util.List | |
| import java.util.Map | |
| import java.util.HashMap | |
| import java.nio.file.Path | |
| import java.nio.file.Files |
| SELECT DISTINCT ?entity ?entityLabel ?propertyLabel ?valueLabel ?retracted ?retractedLabel WITH { | |
| SELECT DISTINCT ?entity ?p ?statement ?retracted WHERE { | |
| ?retracted wdt:P31 wd:Q45182324 . | |
| ?entity ?p ?statement . | |
| ?statement prov:wasDerivedFrom / pr:P248 ?retracted . | |
| } | |
| } AS %STATEMENTS WHERE { | |
| INCLUDE %STATEMENTS | |
| ?p ^wikibase:claim ?property . | |
| ?statement ?ps ?value . |
| Pathways without any tag: 74. Details: | |
| http://classic.wikipathways.org/instance/WP3104_r118203 'Mevalonate pathway (Arabidopsis thaliana)' | |
| http://classic.wikipathways.org/instance/WP3029_r118242 'Vitamin E biosynthesis (Arabidopsis thaliana)' | |
| http://classic.wikipathways.org/instance/WP3113_r118262 'Gibberellin biosynthesis III (early C-13 hydroxylation) (Arabidopsis thaliana)' | |
| http://classic.wikipathways.org/instance/WP3024_r116502 'Formation of ternary complex, then 43S complex (Oryza sativa)' | |
| http://classic.wikipathways.org/instance/WP3066_r118243 'Glutamine biosynthesis I (Arabidopsis thaliana)' | |
| http://classic.wikipathways.org/instance/WP3083_r118610 'Gamma-glutamyl cycle (Zea mays)' | |
| http://classic.wikipathways.org/instance/WP3003_r118295 'Cap-dependent translation initiation (Oryza sativa)' | |
| http://classic.wikipathways.org/instance/WP3087_r118265 'GDP-L-fucose biosynthesis I from GDP-D-mannose (Arabidopsis thaliana)' | |
| http://classic.wikipathways.org/instance/WP3035_r118280 'IAA biosynthesis VI via i |
| PREFIX target: <http://www.wikidata.org/entity/Q76> | |
| SELECT DISTINCT ?source ?sourceLabel WHERE { | |
| VALUES ?refType { pr:P248 pr:P854 } | |
| target: ?propp ?statement . | |
| ?statement prov:wasDerivedFrom ?reference . | |
| ?reference ?refType ?source . | |
| SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". } | |
| } |
| SELECT DISTINCT ?np ?chembl WHERE { | |
| ?np wdt:P703 [] ; wdt:P592 ?chembl . | |
| } |
Some files related to some data curation I did yesterday to add some 200~ish CiTO annotations to Wikidata. The data is CCZero, from https://doi.org/10.5281/zenodo.6885109
There are two files here:
cits_cleaned.csv derived from the cits.csv from the Zenodo dataquickstatements.groovy that takes cits_cleaned.csv as input and creates QuickStatements