Package index • bioclients

Gene identity

Where nearly every other lookup starts, because the rest are keyed on an Entrez, Ensembl, UniProt, or HGNC id rather than a symbol.

mygene_gene(): Look up one gene
mygene_genes(): Look up many genes in one request
mygene_parse_hits(): Turn MyGene hits into a gene table
mygene_parse_batch(): Turn a MyGene batch response into a gene table
mygene_pick_hit(): Choose the best MyGene hit for a queried token

gnomAD

Population frequency and gene constraint, as separate entry points. One function cannot serve both.

gnomad_constraint(): Gene constraint for one gene
gnomad_constraints(): Gene constraint for many genes
gnomad_frequency(): Population allele frequency for a variant
gnomad_parse_constraint(): Turn a gnomAD constraint response into a table
gnomad_parse_constraints(): Turn an aliased gnomAD constraint response into a table
gnomad_parse_frequency(): Turn a gnomAD variant response into a frequency record
gnomad_parse_populations(): Combine gnomAD per-ancestry counts into one frequency table

ClinVar

Clinical significance, via NCBI E-utilities.

clinvar_classification(): Look up the ClinVar classification for a variant
clinvar_parse_record(): Turn a ClinVar esummary record into a table
clinvar_category(): Bucket a ClinVar significance string into a coarse category
clinvar_conditions(): Collapse a ClinVar trait set into one condition string

MyVariant

Precomputed variant annotation, 1000 per batch. The request always carries assembly = "hg38"; without it GRCh38 coordinates come back notfound.

myvariant_variants(): Annotate many variants in one request
myvariant_parse_record(): Turn one MyVariant record into a table row
myvariant_parse_batch(): Turn a MyVariant batch response into a table
myvariant_id(): Build a MyVariant identifier from variant components

Ensembl VEP

Predicted consequence for genomic coordinates, exactly 200 per POST. Results are matched back by identity, because VEP does not promise order.

vep_variants(): Consequence predictions for many variants
vep_parse_element(): Turn one VEP element into a table row
vep_parse_batch(): Turn a VEP batch response into a table
vep_pick_transcript(): Choose which transcript to report for a variant
vep_key(): Build the variant key VEP results are matched on
vep_region(): Build a VEP region string from variant components

Ensembl REST

VEP by variant id, and the exon model. Exons are numbered in transcription order, so on the minus strand exon 1 has the highest coordinate.

ensembl_vep_id(): Run VEP for a variant id
ensembl_gene_model(): The exon model for a gene
ensembl_parse_vep(): Turn an Ensembl VEP record into a result
ensembl_parse_consequences(): Turn Ensembl transcript consequences into a table
ensembl_parse_gene_model(): Turn an Ensembl gene lookup into a gene model

VariantValidator

HGVS normalisation. The response is keyed by the normalised variant, which is not known before the call.

variantvalidator_normalize(): Validate and normalize one HGVS variant
variantvalidator_parse(): Turn a VariantValidator response into a table

ClinGen Allele Registry

Canonical allele identifiers, as a keyless batch POST.

clingen_alleles(): Resolve HGVS to canonical allele ids
clingen_parse_allele(): Turn one Allele Registry element into a table row
clingen_parse_batch(): Turn an Allele Registry batch response into a table

Open Targets

Gene and disease association in both directions, the disease resolver that existed twice in the family, known drugs, and pharmacogenomics.

opentargets_gene_diseases(): Diseases associated with a gene
opentargets_disease_targets(): Genes associated with a disease
opentargets_resolve_disease(): Resolve a disease term to ontology records
opentargets_drugs(): Known drugs and clinical candidates for a gene
opentargets_pgx(): Pharmacogenomics annotations for a gene
opentargets_parse_diseases(): Turn target-to-disease rows into a table
opentargets_parse_targets(): Turn disease-to-target rows into a table
opentargets_parse_matches(): Turn a disease search or lookup into a table
opentargets_parse_drugs(): Turn known-drug rows into a table
opentargets_parse_pgx(): Turn pharmacogenomics rows into a table
opentargets_is_id(): Is a term an ontology id rather than free text

DGIdb

Curated drug-gene interaction counts. An unknown gene is NA, a known gene with none is 0.

dgidb_gene(): Drug-gene interaction count for one gene
dgidb_genes(): Drug-gene interaction counts for many genes
dgidb_parse_genes(): Turn a DGIdb genes response into a table

Pharos

Target Development Level. The classification only; turning it into a weight is the caller’s job.

pharos_target(): Target Development Level for one gene
pharos_targets(): Target Development Level for many genes
pharos_parse_targets(): Turn a Pharos targets response into a table

CIViC

Expert-curated cancer variant evidence counts.

civic_gene(): Curated clinical evidence counts for a gene
civic_parse_gene(): Turn a CIViC gene response into a table

JensenLab DISEASES

Disease-to-gene associations scored 0 to 5, in a curated and a text-mined channel. Combining them is a convention, not the source’s own answer.

diseases_channel(): Genes DISEASES associates with a disease, from one channel
diseases_gene_associations(): Genes DISEASES associates with a disease, across both channels
diseases_parse_channel(): Turn one DISEASES channel response into a table
diseases_merge_channels(): Combine DISEASES channels, keeping the strongest score per gene

PanelApp

Diagnostic gene panels. The traffic-light level is returned as published, red included, rather than as a weight.

panelapp_panels(): One page of the PanelApp panel index
panelapp_all_panels(): The whole PanelApp panel index
panelapp_panel(): The genes on one PanelApp panel
panelapp_parse_index(): Turn a PanelApp panel index page into a table
panelapp_parse_panel(): Turn a PanelApp panel detail into a table of genes

ClinGen gene validity

Curated gene-disease validity. There is no per-gene endpoint, so the bulk file is fetched once and filtered in memory.

clingen_gene_validity(): The ClinGen gene-disease validity table
clingen_parse_validity(): Parse the ClinGen gene-validity CSV
clingen_validity_for(): Filter a parsed validity table to one or more genes

UniProt

Disease annotation and sequence features. Two hosts, kept apart so each gets its own circuit breaker.

uniprot_diseases(): Diseases UniProt curates for an accession
uniprot_features(): Sequence features for an accession
uniprot_parse_diseases(): Turn a UniProtKB entry into a curated-disease table
uniprot_parse_features(): Turn an EBI Proteins features response into a table
uniprot_features_at(): The features spanning a residue position

ProtVar

Residue-level function and population observations for a protein position.

protvar_function(): Functional context for a residue
protvar_population(): Known variants at a residue
protvar_parse_function(): Turn a ProtVar function response into its text
protvar_parse_population(): Turn a ProtVar population response into a table
protvar_position(): Pull a residue position out of a protein-change string
protvar_strip_citations(): Strip inline citations out of a UniProt function comment

AlphaFold

Predicted structure and its per-residue confidence.

alphafold_model(): The predicted structure for a UniProt accession
alphafold_parse_model(): Turn an AlphaFold prediction response into a table

PDBe

Experimental structures covering a protein, collapsed to one row per structure rather than one per chain.

pdbe_structures(): Experimental structures for a UniProt accession
pdbe_parse_structures(): Turn a PDBe best-structures response into a table

STRING

Interaction partners and networks. Returned names are translated back to the symbols that were queried.

string_partners(): Interaction partners for one gene
string_network(): The interaction network within a set of genes
string_map_ids(): The STRING identifier map for a set of symbols
string_parse_partners(): Turn STRING interaction-partner rows into a table
string_parse_network(): Turn a STRING network response into an edge table
string_parse_ids(): Turn a STRING identifier-map response into a table
string_reconcile_edges(): Rewrite edge endpoints back into the queried symbol space

GTEx

Median expression across tissues, which needs GTEx’s own versioned GENCODE id first. An Ensembl id from elsewhere will not work.

gtex_gene_reference(): Resolve a gene to GTEx's versioned GENCODE id
gtex_median_expression(): Median expression across tissues
gtex_parse_reference(): Turn a GTEx gene-reference response into a table
gtex_parse_expression(): Turn a GTEx median-expression response into a table

Human Protein Atlas

Curated per-gene annotation, returned as the tags HPA publishes.

hpa_gene(): The Human Protein Atlas record for a gene
hpa_parse_gene(): Turn an HPA gene record into a table

QuickGO

Gene Ontology annotations, collapsed to one row per term rather than one per evidence line.

quickgo_annotations(): GO annotations for a UniProt accession
quickgo_parse_annotations(): Turn a QuickGO annotation response into a table

Reactome

Pathway membership, keyed by gene symbol. The response is a bare JSON array rather than an object.

reactome_pathways(): Reactome pathways for a gene symbol
reactome_parse_pathways(): Turn a Reactome pathway array into a table

Human Phenotype Ontology

Term search, term lookup, and the diseases and phenotypes HPO associates with a gene. The gene lookup is keyed on an NCBI Gene id.

hpo_search(): Search HPO terms by free text
hpo_term(): Resolve one HP id to its term
hpo_gene_annotation(): HPO's annotation for a gene
hpo_parse_search(): Turn an HPO search response into a table
hpo_parse_term(): Turn an HPO term response into a table
hpo_parse_diseases(): Turn an HPO gene annotation into a table of diseases
hpo_parse_phenotypes(): Turn an HPO gene annotation into a table of phenotypes

Monarch Initiative

Entity search and the associations between entities. A gene symbol is not unique across species, so every search row carries its taxon.

monarch_search(): Search Monarch for an entity
monarch_associations(): Associations with an entity on one end
monarch_gene_phenotypes(): HPO phenotypes Monarch associates with a gene
monarch_parse_search(): Turn a Monarch search response into a table
monarch_parse_associations(): Turn Monarch association records into a table
monarch_hgnc_id(): Normalise an HGNC id to the CURIE form Monarch expects

IMPC

Mouse knockout phenotypes. A gene never phenotyped and one phenotyped with no significant result are indistinguishable, so neither is reported as 0.

impc_mouse_ortholog(): The mouse ortholog IMPC holds for a human gene
impc_gene_phenotypes(): Significant knockout phenotypes IMPC records for a human gene
impc_parse_ortholog(): Read the mouse ortholog out of an IMPC gene-core response
impc_parse_phenotypes(): Turn an IMPC phenotype response into a table

Europe PMC

Literature search and hit counts. A count of 0 is an answer and comes back ok, because a gene with no literature is not an outage.

europepmc_search(): Search Europe PMC
europepmc_count(): How many publications Europe PMC has for a query
europepmc_query(): Build a Europe PMC query from terms
europepmc_parse_results(): Turn Europe PMC search results into a table
europepmc_parse_count(): Read the hit count off a Europe PMC response

PubTator3

Articles where a gene is tagged as an entity rather than matched as text, so an ambiguous symbol stops inflating its own count.

pubtator_gene_literature(): Articles PubTator3 has tagged with a gene
pubtator_entity(): Build the PubTator3 entity token for a gene
pubtator_parse_count(): Read the article count off a PubTator3 search response
pubtator_parse_results(): Turn PubTator3 search results into a table

Package

bioclients bioclients-package: bioclients: Clients for Biological Database Web Services