The goal of graphql.parse is to parse GraphQL schemas in an R-style way
You can install the development version of graphql.parse like so:
remotes::install_github("HIBio/graphql.parse")query_builder() walks up and down the schema, enabling users to
dynamically select the fields they wish to be returned
library(graphql.parse)
query_string <- query_builder()
Options in Query; Select next level, or 0 to finish
1: meta 2: target 3: targets
4: disease 5: diseases 6: drug
7: drugs 8: search 9: associationDatasources
10: interactionResources 11: geneOntologyTerms 12: up 1 level
Selection: 6
Options in Drug; Select next level, or 0 to finish
1: id 2: name 3: synonyms
4: tradeNames 5: yearOfFirstApproval 6: drugType
7: isApproved 8: crossReferences 9: maximumClinicalTrialPhase
10: hasBeenWithdrawn 11: blackBoxWarning 12: description
13: parentMolecule 14: childMolecules 15: approvedIndications
16: drugWarnings 17: similarEntities 18: literatureOcurrences
19: mechanismsOfAction 20: indications 21: knownDrugs
22: adverseEvents 23: linkedDiseases 24: linkedTargets
25: up 1 level
Selection: 1
Options in Drug; Select next level, or 0 to finish
1: id 2: name 3: synonyms
4: tradeNames 5: yearOfFirstApproval 6: drugType
7: isApproved 8: crossReferences 9: maximumClinicalTrialPhase
10: hasBeenWithdrawn 11: blackBoxWarning 12: description
13: parentMolecule 14: childMolecules 15: approvedIndications
16: drugWarnings 17: similarEntities 18: literatureOcurrences
19: mechanismsOfAction 20: indications 21: knownDrugs
22: adverseEvents 23: linkedDiseases 24: linkedTargets
25: up 1 level
Selection: 19
Options in MechanismsOfAction; Select next level, or 0 to finish
1: rows
2: uniqueActionTypes
3: uniqueTargetTypes
4: up 1 level
Selection: 1
Options in MechanismOfActionRow; Select next level, or 0 to finish
1: mechanismOfAction
2: actionType
3: targetName
4: references
5: targets
6: up 1 level
Selection: 1
Options in MechanismOfActionRow; Select next level, or 0 to finish
1: mechanismOfAction
2: actionType
3: targetName
4: references
5: targets
6: up 1 level
Selection: 3
Options in MechanismOfActionRow; Select next level, or 0 to finish
1: mechanismOfAction
2: actionType
3: targetName
4: references
5: targets
6: up 1 level
Selection: 5
Options in Target; Select next level, or 0 to finish
1: id 2: alternativeGenes 3: approvedSymbol 4: approvedName
5: biotype 6: chemicalProbes 7: dbXrefs 8: functionDescriptions
9: geneticConstraint 10: genomicLocation 11: geneOntology 12: hallmarks
13: homologues 14: pathways 15: proteinIds 16: safetyLiabilities
17: subcellularLocations 18: synonyms 19: symbolSynonyms 20: nameSynonyms
21: obsoleteSymbols 22: obsoleteNames 23: targetClass 24: tep
25: tractability 26: transcriptIds 27: similarEntities 28: literatureOcurrences
29: evidences 30: interactions 31: mousePhenotypes 32: expressions
33: knownDrugs 34: associatedDiseases 35: up 1 level
Selection: 1
Options in Target; Select next level, or 0 to finish
1: id 2: alternativeGenes 3: approvedSymbol 4: approvedName
5: biotype 6: chemicalProbes 7: dbXrefs 8: functionDescriptions
9: geneticConstraint 10: genomicLocation 11: geneOntology 12: hallmarks
13: homologues 14: pathways 15: proteinIds 16: safetyLiabilities
17: subcellularLocations 18: synonyms 19: symbolSynonyms 20: nameSynonyms
21: obsoleteSymbols 22: obsoleteNames 23: targetClass 24: tep
25: tractability 26: transcriptIds 27: similarEntities 28: literatureOcurrences
29: evidences 30: interactions 31: mousePhenotypes 32: expressions
33: knownDrugs 34: associatedDiseases 35: up 1 level
Selection: 3
Options in Target; Select next level, or 0 to finish
1: id 2: alternativeGenes 3: approvedSymbol 4: approvedName
5: biotype 6: chemicalProbes 7: dbXrefs 8: functionDescriptions
9: geneticConstraint 10: genomicLocation 11: geneOntology 12: hallmarks
13: homologues 14: pathways 15: proteinIds 16: safetyLiabilities
17: subcellularLocations 18: synonyms 19: symbolSynonyms 20: nameSynonyms
21: obsoleteSymbols 22: obsoleteNames 23: targetClass 24: tep
25: tractability 26: transcriptIds 27: similarEntities 28: literatureOcurrences
29: evidences 30: interactions 31: mousePhenotypes 32: expressions
33: knownDrugs 34: associatedDiseases 35: up 1 level
Selection: 35
Options in Target; Select next level, or 0 to finish
1: mechanismOfAction
2: actionType
3: targetName
4: references
5: targets
6: up 1 level
Selection: 4
Options in Reference; Select next level, or 0 to finish
1: ids
2: source
3: urls
4: up 1 level
Selection: 2
Options in Reference; Select next level, or 0 to finish
1: ids
2: source
3: urls
4: up 1 level
Selection: 3
Options in Reference; Select next level, or 0 to finish
1: ids
2: source
3: urls
4: up 1 level
Selection: 0
query gql(
$chemblId: String!
) {
drug(chemblId: $chemblId) {
id
mechanismsOfAction {
rows {
mechanismOfAction
targetName
targets {
id
approvedSymbol
}
references {
source
urls
}
}
}
}
}
Template for variables:
variables = list(
chemblId = character(1)
)
This query_string can then be passed along with the necessary
variables to the API
run_query(query_string, variables = list(chemblId = "CHEMBL1201828"))
#> $drug
#> $drug$id
#> [1] "CHEMBL1201828"
#>
#> $drug$mechanismsOfAction
#> $drug$mechanismsOfAction$rows
#> $drug$mechanismsOfAction$rows[[1]]
#> $drug$mechanismsOfAction$rows[[1]]$mechanismOfAction
#> [1] "Complement C5 inhibitor"
#>
#> $drug$mechanismsOfAction$rows[[1]]$targetName
#> [1] "Complement C5"
#>
#> $drug$mechanismsOfAction$rows[[1]]$targets
#> $drug$mechanismsOfAction$rows[[1]]$targets[[1]]
#> $drug$mechanismsOfAction$rows[[1]]$targets[[1]]$id
#> [1] "ENSG00000106804"
#>
#> $drug$mechanismsOfAction$rows[[1]]$targets[[1]]$approvedSymbol
#> [1] "C5"
#>
#>
#>
#> $drug$mechanismsOfAction$rows[[1]]$references
#> $drug$mechanismsOfAction$rows[[1]]$references[[1]]
#> $drug$mechanismsOfAction$rows[[1]]$references[[1]]$source
#> [1] "DailyMed"
#>
#> $drug$mechanismsOfAction$rows[[1]]$references[[1]]$urls
#> $drug$mechanismsOfAction$rows[[1]]$references[[1]]$urls[[1]]
#> [1] "http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=ebcd67fa-b4d1-4a22-b33d-ee8bf6b9c722"The schema can also be fetched once then re-used in additional queries
schema <- get_schema()
cat(substring(schema, 1, 50))
#> type APIVersion {
#> x: String!
#> y: String!
#> z: SThis can then be used in the query_builder() without hitting the
endpoint just to retrieve the schema
query_builder(schema = schema)Alternatively, the Genetics API can be used. When an api_url is
specified in query_builder(), it is stored in the query string and
accessible to run_query
qs <- query_builder(api_url = OT_GENETICS_API)Options in Query; Select next level, or 0 to finish
1: meta 2: search
3: genes 4: geneInfo
5: studyInfo 6: variantInfo
7: studiesForGene 8: studyLocus2GeneTable
9: manhattan 10: topOverlappedStudies
11: overlapInfoForStudy 12: tagVariantsAndStudiesForIndexVariant
13: indexVariantsAndStudiesForTagVariant 14: pheWAS
15: gecko 16: regionPlot
17: genesForVariantSchema 18: genesForVariant
19: gwasRegional 20: qtlRegional
21: studyAndLeadVariantInfo 22: gwasCredibleSet
23: qtlCredibleSet 24: colocalisationsForGene
25: gwasColocalisationForRegion 26: gwasColocalisation
27: qtlColocalisation 28: studiesAndLeadVariantsForGene
29: studiesAndLeadVariantsForGeneByL2G 30: up 1 level
Selection: 4
Options in Gene; Select next level, or 0 to finish
1: id 2: symbol 3: bioType 4: description 5: chromosome 6: tss
7: start 8: end 9: fwdStrand 10: exons 11: up 1 level
Selection: 1
Options in Gene; Select next level, or 0 to finish
1: id 2: symbol 3: bioType 4: description 5: chromosome 6: tss
7: start 8: end 9: fwdStrand 10: exons 11: up 1 level
Selection: 2
Options in Gene; Select next level, or 0 to finish
1: id 2: symbol 3: bioType 4: description 5: chromosome 6: tss
7: start 8: end 9: fwdStrand 10: exons 11: up 1 level
Selection: 4
Options in Gene; Select next level, or 0 to finish
1: id 2: symbol 3: bioType 4: description 5: chromosome 6: tss
7: start 8: end 9: fwdStrand 10: exons 11: up 1 level
Selection: 0
query gql(
$geneId: String!
) {
geneInfo(geneId: $geneId) {
id
symbol
description
}
}
Template for variables:
variables = list(
geneId = character(1)
)
run_query(qs, variables = list(geneId = "ENSG00000197405"))
#> $geneInfo
#> $geneInfo$id
#> [1] "ENSG00000197405"
#>
#> $geneInfo$symbol
#> [1] "C5AR1"
#>
#> $geneInfo$description
#> [1] "complement C5a receptor 1 [Source:HGNC Symbol;Acc:HGNC:1338]"Some canned queries are available. To fetch the dataVersion
dataVersion(OT_API)
#> Warning in api_url.default(query_string): No api_url detected on this object
#> $name
#> [1] "Open Targets GraphQL & REST API Beta"
#>
#> $dataVersion
#> year month iteration
#> "23" "09" "0"
dataVersion(OT_GENETICS_API)
#> Warning in api_url.default(query_string): No api_url detected on this object
#> $name
#> [1] "Open Targets Genetics Portal"
#>
#> $dataVersion
#> major minor patch
#> 22 10 0This video provides a great overview of the OpenTargets GraphQL API.
This package is designed to work with any GraphQL schema, but is
currently limited to those endpoints which support a plaintext schema
hosted at <GRAPHQL_API>/schema. An improvement to this package would
involve fetching the schema itself via the API.
The OpenTargets (and OpenTargets Genetics) GraphQL APIs are currently
supported and the APIs are stored as exported variables in this package,
OT_API and OT_GENETICS_API.