Gene

Use gene commands to retrieve canonical metadata and targeted biological context.

What the gene guide covers

• symbol-based retrieval,
• lightweight search,
• section expansion,
• JSON output for downstream systems.

Search genes

Start with search when you are unsure of symbol spelling or want to inspect alias candidates. Known aliases that map to one canonical human gene can also be passed directly to get gene.

biomcp search gene BRAF --limit 5

Useful fields in search output typically include symbol, Entrez ID, and species.

Get a gene record

biomcp get gene BRAF

The default gene view is concise and intended for orientation. When MyGene.info returns genomic coordinates, BioMCP labels them with the GRCh38 genome build (Chromosome (GRCh38): ...) so coordinate consumers do not have to infer the reference. Its More: block keeps pathways, ontology, and diseases visible and now also surfaces funding as a direct follow-up from the base card.

Request deeper sections

BioMCP expands detail via positional sections.

Pathway view:

biomcp get gene BRAF pathways

Disease associations:

biomcp get gene BRAF diseases

Ontology terms:

biomcp get gene BRAF ontology

Protein summary:

biomcp get gene BRAF protein

When UniProt exposes legacy protein names, the protein section includes an Also known as: line with alternative full names and short names from UniProt. When UniProt exposes alternative products, the same section also includes an Isoforms (N) line with isoform names and the displayed isoform length when that length is available from the base UniProt record.

GO terms and interactions:

biomcp get gene BRAF go interactions

CIViC evidence summary:

biomcp get gene BRAF civic

Tissue expression (GTEx):

biomcp get gene BRAF expression

Protein tissue expression and localization (Human Protein Atlas):

biomcp get gene BRAF hpa

Druggability profile (DGIdb interactions plus OpenTargets tractability and safety):

biomcp get gene BRAF druggability

Funding context (NIH Reporter grants mentioning the canonical symbol in the most recent 5 NIH fiscal years):

biomcp get gene ERBB2 funding

Diagnostic-test pivot (GTR tests for the gene):

biomcp get gene BRCA1 diagnostics

The diagnostics and funding sections are opt-in and are not included in biomcp get gene <symbol> all.

Gene-disease validity (ClinGen):

biomcp get gene BRAF clingen

The additive GeneClinGen JSON shape keeps the existing evidence fields and reports validity and dosage acquisition independently:

{
  "validity": [{"disease": "Li-Fraumeni syndrome", "classification": "Definitive"}],
  "haploinsufficiency": "Sufficient Evidence for Haploinsufficiency",
  "triplosensitivity": "No Evidence for Triplosensitivity",
  "validity_status": {"status": "data", "op": "gene_validity_download"},
  "dosage_status": {"status": "data", "op": "gene_dosage_download"}
}

Each family status is one of data, empty, failed, or timed_out; its operation is one of client_init, gene_lookup, gene_validity_download, or gene_dosage_download. Healthy statuses omit message. Failures and timeouts include a stable public message without provider bodies, URLs, paths, or parser details. A failed lookup does not erase an exact-symbol match, but it prevents a zero match from being reported as confirmed empty.

The combined section_outcomes.clingen and _meta.section_sources entry use data when both families are healthy and either has data, and empty only when both are confirmed empty. Data plus an unavailable family is degraded with ClinGen source credit; without any data, a failed or timed-out family is unavailable with no source credit.

Missing dosage fields remain absent in JSON and Markdown. They are not rendered as No evidence; a literal ClinGen classification such as No Evidence for Triplosensitivity is real data and is preserved verbatim.

When ClinGen validity rows are present, the gene card's See also: block promotes a recruiting-trial search keyed to the newest reviewed disease label already shown on the card, ahead of the generic gene pivots.

ClinGen CSpec source documents use a separate, versioned retrieval flow. List a gene's returned resource IRIs, select one exact IRI or a unique short version, then use its capture handle to stream the original stored bytes locally. The display version is not interchangeable with resource identity:

biomcp --json gene cspec ATM
biomcp --json gene cspec ATM --version https://cspec.genome.network/cspec/SequenceVariantInterpretation/id/GN020/version/1.5.1
biomcp --json gene cspec ATM --version 1.5.1
biomcp gene cspec document <capture-id>
biomcp --json gene cspec PTEN --version <full-resource-iri> --files
biomcp --json gene cspec PTEN --capture-id <capture-id> --files

CSpec returns source facts and provenance; it does not evaluate ACMG criteria or classify variants. The opt-in files view lists bounded metadata for linked public attachments without downloading them. Normal criteria output reports attachment_count; capture-based file listing never refetches the provider.

Constraint metrics (gnomAD):

biomcp get gene BRAF constraint

Multiple sections can be chained:

biomcp get gene BRAF pathways diseases

All supported sections:

biomcp get gene BRAF all

To add funding, request it explicitly:

biomcp get gene BRAF funding

Helper commands

biomcp gene trials BRAF --limit 5
biomcp gene trials TP53 --limit 5
biomcp gene drugs BRAF --limit 5
biomcp gene pathways BRAF
biomcp gene articles BRAF
biomcp gene definition BRAF
biomcp gene cell-lines FLT3 --group leukemia

Gene trial pivots send the supplied symbol as a biomarker.

gene cell-lines prints the Human Protein Atlas RNA level (nTPM) of one gene in every HPA cell line of one cancer group, as published. Each row carries the Cellosaurus accession when exactly one human cell line carries the HPA name, and - otherwise. See Human Protein Atlas for the 30 group names and Cell line for the accession the rows join on.

Common workflows

Clinical trial pivot

biomcp search trial -c melanoma --mutation "BRAF V600E" --limit 5

Literature pivot

biomcp search article -g BRAF -d melanoma --limit 5

Variant pivot

biomcp search variant -g BRAF --limit 5

Error handling expectations

If a section name is unsupported, BioMCP returns an explicit unknown-section message with hints about valid section names.

JSON mode

Use JSON for pipelines or agent post-processing.

biomcp --json get gene BRAF

biomcp --json get gene BRAF druggability includes DGIdb interaction fields plus OpenTargets tractability[] modality summaries and safety_liabilities[] event summaries.

Optional-section outcomes

JSON and MCP gene records include all 15 optional keys under section_outcomes. Requested sections such as go and interactions report data, empty, degraded, or unavailable; unrequested keys remain not_requested. An empty payload is therefore a confirmed zero only when its outcome is empty. Markdown prints an in-band status note for unavailable or partial sections.

Practical tips

• Keep section requests narrow for better focus.
• Start with one section, then add another only if needed.
• Use search first when symbol ambiguity is possible.

Related guides

• Variant
• Article
• Trial
• Protein
• Progressive disclosure