Scientific databases

The app bundles 23 data-source Connectors, plus a separate offline Molecule Connector. The full registry has 239 tool operations including Molecule's two operations; the data-source catalog below covers 237. Enable the relevant Connector in Settings, then ask a bounded question with the correct identifier type.

Data-source catalog

Choose by identifier and research question. Source coverage differs; consult the operation reference for exact fields.

Connector	Sources	Operations	Use it for
Chemistry · chemistry	PubChem, ChEBI, Rhea, BindingDB	12	Small-molecule chemistry via PubChem, ChEBI, Rhea and BindingDB.
Literature Graph · literature	OpenAlex, arXiv, Crossref, DataCite	13	Papers, authors, citations, DOI updates and dataset/software records.
PubMed · pubmed	PubMed, PMC, Europe PMC	7	Biomedical literature via NCBI E-utilities, the PMC ID Converter and Europe PMC — search, metadata, related articles, citation lookup, ID conversion, full text and copyright.
Genes & Ontologies · genes	MyGene, UniProt, OLS, QuickGO, Reactome	7	Gene/protein identity and ontology terms — mygene.info, UniProt, OLS4 ontologies, GO annotations, Reactome pathways.
Genomes · genomes	Ensembl, UCSC	11	Genome annotation, variants, homology, sequence and browser tracks — Ensembl REST and the UCSC Genome Browser.
Variants · variants	gnomAD, ClinVar, dbSNP	15	Human genetic variants — gnomAD population frequencies/constraint, ClinVar records/search (direct NCBI), dbSNP, structural and mitochondrial variants.
Clinical Trials · clinical-trials	ClinicalTrials.gov	6	Clinical trials from ClinicalTrials.gov — search, details, sponsors, investigators, endpoints, and eligibility.
Clinical Genomics · clinical-genomics	ClinGen, CIViC, Open Targets	20	Clinical genomics knowledge bases: ClinGen curations, CIViC clinical evidence, and the Open Targets Platform.
Structures & Interactions · structures	PDB, AlphaFold, EMDB, Complex Portal, IntAct	16	Structures and molecular interactions — PDB structures, AlphaFold predictions, EMDB cryo-EM entries, Complex Portal complexes, IntAct interaction networks.
ChEMBL · chembl	ChEMBL	6	Bioactive compounds, drugs, targets, bioactivity, and mechanisms via the ChEMBL REST API.
bioRxiv · biorxiv	bioRxiv, medRxiv, ROR	7	bioRxiv/medRxiv preprints — search by date/category, metadata by DOI, journal-publication links, funder listings, and platform statistics.
Drug Regulatory · drug-regulatory	openFDA	7	Drugs@FDA applications, labels, and corpus statistics via openFDA.
Human Genetics · human-genetics	GWAS Catalog, eQTL Catalogue, PheWeb	14	Human genetic association evidence — GWAS Catalog, eQTL Catalogue, and PheWeb PheWAS portals (FinnGen, BioBank Japan).
Expression · expression	GTEx	12	Human tissue expression and eQTLs via the GTEx Portal.
Protein Annotation · protein-annotation	InterPro, Pfam, Human Protein Atlas, STRING	13	Protein domain architecture, family/clan membership, expression atlas and interaction networks via InterPro/Pfam, the Human Protein Atlas and STRING.
Cancer Models · cancer-models	cBioPortal	6	Cancer genomics study records via the cBioPortal REST API.
RNA · rna	Rfam	9	Non-coding RNA family data (metadata, alignments, models, structures) via Rfam.
Omics Archives · omics-archives	ArrayExpress, GEO, MetaboLights, MGnify, PRIDE	17	Omics data archives — expression (ArrayExpress, GEO), metabolomics (MetaboLights), metagenomics (MGnify) and proteomics (PRIDE).
CellGuide · cellguide	CELLxGENE	5	Cell-type identity, marker genes, source datasets, and tissues via CELLxGENE CellGuide.
Regulation · regulation	ENCODE, JASPAR, UniBind	16	Gene-regulation functional genomics — ENCODE experiments/biosamples/files, JASPAR TF binding profiles, and UniBind ChIP-seq TFBS.
Research Resources · research-resources	Grants.gov, Antibody Registry	5	Funding-opportunity search (Grants.gov) and antibody catalog lookups (Antibody Registry).
BioMart · biomart	Ensembl BioMart	8	Ensembl BioMart attribute queries and identifier translation.
ZINC · zinc	ZINC	5	ZINC22 purchasable chemical space (CartBlanche22) — compound lookup by ZINC id, SMILES exact/similarity search, supplier-code resolution, random sampling, 3D structure locations for docking.

Retrieve a record and verify its identity

1. Open Settings → Connectors, search the required source and confirm availability to the intended agent.
2. Open its detail. Read Tools, inputs, example and third-party requirements.
3. Supply an explicit query/accession and result limit. Retain the exact query when creating a literature collection or evidence table.
4. Inspect returned IDs and source fields. An empty result, a truncated batch and an error are different outcomes.
5. Save needed records into the project/library deliberately. A search response does not automatically mean all papers were added to the literature library or full texts downloaded.

Start with one known identifier

Worked example Resolve the human TP53 gene identifier

Enable Genes & Ontologies and ask: Use query_genes to resolve TP53 with scopes="symbol", species="human" and fields="symbol,name,entrezgene". Return the input query and any unmatched records. In this example, the human TP53 record identifies Entrez Gene 7157 and the name tumor protein p53. Check the record's query and symbol before using the mapped ID. A symbol may return several matches, so retain all results until you have confirmed the intended organism and record. Exact fields.

Choose a query and inspect the result

Example Bounded database queries and responses

The table records these example responses; live query results may differ.

Connector / tool	Input	Observed result
Omics Archives / geo_get_series	accessions: ["GSE60450"]	Series/sample metadata with 12 samples; metadata retrieval did not recompute the uploaded counts.
Genes / query_genes	TP53; symbol scope; human	Entrez Gene ID 7157, symbol TP53, name tumor protein p53.
PubMed / search_articles	GSE60450, maximum 2	PMIDs 38059347 and 37306301. These are query matches, not automatically the dataset's original publication.
Chemistry / pubchem_search_compounds	aspirin, maximum 1 CID	CID 2244, formula C9H8O4 and molecular weight 180.16.
Literature / openalex_search_works	CRISPR base editing; from 2020; open access; maximum 2	Two work records with OpenAlex IDs, source fields and completeness flags.

Connect OpenAlex and follow citation links

1. Open Settings → Connectors → Literature Graph → Manage credentials → OpenAlex.
2. Enter your API key, select Validate, then Save after validation succeeds.
3. Search for a topic with a small max_records limit. Check n_records_returned and records_truncated before describing the result as complete.
4. Use a returned work ID with openalex_get_work. Use openalex_citations for papers citing that work and openalex_references for works it cites. These are opposite directions.
5. For author searches, confirm the institution and ORCID before retrieving an author profile. Use a source ID or ISSN to disambiguate a journal name.

See the OpenAlex operation parameters for filters and returned fields.

Look up a DOI and its related research records

Enable Literature Graph. Use crossref_get_work for publisher metadata and crossref_get_updates for deposited correction/retraction relationships. Use datacite_search_records to find dataset/software DOIs, then datacite_get_record to inspect a selected record. These four public methods do not require the OpenAlex key. Verify DOI identity, relationship direction and reuse terms before downloading or citing a resource. Exact fields are in the operation reference.

Rfam sequence search now uses the official batch endpoint. If an older installation returns the retired-endpoint error, update the app and retry the intended operation. A pending job is not a completed search with no hits.

Handle a returned record, empty match or error

Inspect the returned status before using a result. Use the operation reference to interpret fields and completeness flags.

Observed outcome	What to do next
found: false, zero records, empty investigators or supplier matches	Check identifier, organism, query scope and filters. Preserve the empty result; do not present it as a retrieved record.
credential_required for OpenAlex	Open the requested credential form and bind your own key before retrying.
contact_email_required for direct NCBI variant queries	Open Settings → Connectors → Manage credentials → Literature access, enter Contact email and select Save. Retry the failed query. An NCBI API key is optional. Check returned identifiers, match counts and truncation flags; an empty result is distinct from a connection error.
HTTP 410 from eQTL	Retain the source URL, operation and response, and check service availability before changing scientific inputs.
Connector request timed out after 30000ms	Retry a smaller request. Increasing only the outer Notebook timeout does not change the Connector's own deadline.
Notebook execution timed out after 60000ms	The execution ended without a result. Retry operations individually; do not infer that every upstream service failed.
BioMart HTML maintenance page; PRIDE Unexpected end of JSON input	The expected structured response was unavailable. Retry later and keep the response type/error for an issue.
ZINC task did not complete in time	Preserve the returned task/result URL and check that job; repeatedly starting new jobs does not recover its result.

For a report, attach the operation, bounded input, error text and timestamp through Troubleshooting. Remove credentials and private data before sharing.

Find operation parameters

Use the Connector operation reference for required fields, accepted values and exact calls. Choose a source here first; use the reference when preparing a specific operation.

Keep genome build, organism, tissue, units and accession versions with returned data. For general HTTP meanings and recovery, use Troubleshooting. Database records, predictions and generated summaries are different evidence types; check the cited source before using a research claim.

Implementation reference: ConnectorsPanel.tsx.

Catalog source: catalog.ts, registry.ts.