gnomad-genetics-mcp-server

v0.6.0 pre-1.0

Ancestry allele frequencies, LoF constraint, variant lists, and coverage from gnomAD, joined with ClinVar clinical significance via MCP. STDIO or Streamable HTTP.

gnomad-genetics.caseyjhand.com/mcp
claude mcp add --transport http gnomad-genetics-mcp-server https://gnomad-genetics.caseyjhand.com/mcp
codex mcp add gnomad-genetics-mcp-server --url https://gnomad-genetics.caseyjhand.com/mcp
{
  "mcpServers": {
    "gnomad-genetics-mcp-server": {
      "url": "https://gnomad-genetics.caseyjhand.com/mcp"
    }
  }
}
gemini mcp add --transport http gnomad-genetics-mcp-server https://gnomad-genetics.caseyjhand.com/mcp
{
  "mcpServers": {
    "gnomad-genetics-mcp-server": {
      "command": "bunx",
      "args": [
        "mcp-remote",
        "https://gnomad-genetics.caseyjhand.com/mcp"
      ]
    }
  }
}
{
  "mcpServers": {
    "gnomad-genetics-mcp-server": {
      "type": "http",
      "url": "https://gnomad-genetics.caseyjhand.com/mcp"
    }
  }
}
curl -X POST https://gnomad-genetics.caseyjhand.com/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-11-25","capabilities":{},"clientInfo":{"name":"curl","version":"1.0.0"}}}'

Tools

9

read 8

gnomad_get_variant

open-world

Fetch the full gnomAD population record for one or more variants — allele count/number/frequency overall and broken down per genetic-ancestry group, the filtering allele frequency (faf95) to test against a disease threshold such as ACMG/AMP BA1 or BS1, homozygote and hemizygote counts, quality flags, each callset's own counts and QC filters with a qc_status, the most severe transcript consequence with its HGVS and LOFTEE loss-of-function call (confidence, filter, flags) beside the same on the gene's MANE Select (else canonical) transcript, gnomAD's manual LoF curation verdicts, in-silico predictor scores, and joined ClinVar clinical significance. The "how common, is it benign" answer in one call. A qc_status of filtered means the variant failed gnomAD QC in every callset carrying it (gnomAD's browser hides these by default; AC0 means no high-quality carrier), so treat its counts as low-confidence rather than an observed frequency. Accepts a batch of up to 25 IDs (chrom-pos-ref-alt or rsID) with per-item partial success: a malformed or absent ID lands in failed[] — with its reason and a recovery hint — without failing the others. A coordinate ID carries no build: an absent one whose string also lifts over from the other build fails as variant_not_found_other_build, with the lifted IDs on the dataset's build in candidates, while a string valid on both builds can name a different variant on each, which no not-found path can see — so keep IDs on the dataset's build. variant_not_found means the variant is not in the chosen dataset — pair with gnomad_get_coverage to confirm the position is callable before concluding true absence. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_get_variant",
    "arguments": {
      "variants": "<variants>"
    }
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "variants": {
      "minItems": 1,
      "maxItems": 25,
      "type": "array",
      "items": {
        "type": "string",
        "minLength": 1,
        "description": "Variant ID — chrom-pos-ref-alt (1-based, e.g. 1-55051215-G-GA) on chromosome 1–22, X, or Y with an optional chr prefix, or an rsID (rs11591147). Mitochondrial IDs (M, MT, chrM) are not served. Obtain a variantId from ensembl_predict_variant or a VCF. Malformed IDs are reported per-item in failed[], not rejected wholesale."
      },
      "description": "1–25 variant IDs (chrom-pos-ref-alt or rsID) to look up in one batched call."
    },
    "dataset": {
      "description": "gnomAD dataset, default gnomad_r4: gnomad_r4 and its subset gnomad_r4_non_ukb (GRCh38); gnomad_r3 and its subsets gnomad_r3_controls_and_biobanks, gnomad_r3_non_cancer, gnomad_r3_non_neuro, gnomad_r3_non_topmed, gnomad_r3_non_v2 (GRCh38); gnomad_r2_1 and its subsets gnomad_r2_1_controls, gnomad_r2_1_non_cancer, gnomad_r2_1_non_neuro, gnomad_r2_1_non_topmed (GRCh37); exac (GRCh37). A subset keeps only part of its parent release's samples (non_cancer leaves out cancer-study samples, non_ukb UK Biobank's) and can leave a callset unchanged: gnomad_r4_non_ukb genome counts equal gnomad_r4's, and gnomad_r2_1_non_cancer genome counts equal gnomad_r2_1's. No subset publishes joint data, and gnomAD publishes no subset coverage or constraint, so those come from the parent. Echoed in output.",
      "type": "string",
      "enum": [
        "gnomad_r4",
        "gnomad_r4_non_ukb",
        "gnomad_r3",
        "gnomad_r3_controls_and_biobanks",
        "gnomad_r3_non_cancer",
        "gnomad_r3_non_neuro",
        "gnomad_r3_non_topmed",
        "gnomad_r3_non_v2",
        "gnomad_r2_1",
        "gnomad_r2_1_controls",
        "gnomad_r2_1_non_neuro",
        "gnomad_r2_1_non_cancer",
        "gnomad_r2_1_non_topmed",
        "exac"
      ]
    },
    "reference_genome": {
      "description": "Reference build. Derived from dataset when omitted: GRCh38 for gnomad_r4, gnomad_r3, and their subsets; GRCh37 for gnomad_r2_1, its subsets, and exac. If supplied it must match the dataset, or the call is rejected. Keep aligned with ensembl coordinates.",
      "type": "string",
      "enum": [
        "GRCh38",
        "GRCh37"
      ]
    }
  },
  "required": [
    "variants"
  ],
  "additionalProperties": false
}
view source ↗

gnomad_get_gene_constraint

open-world

Fetch gnomAD loss-of-function constraint for a gene — pLI (probability of LoF intolerance; >0.9 intolerant), LOEUF (oe_lof_upper, the headline metric) plus its lower bound and gnomAD’s LOEUF percentile, observed/expected ratios for LoF, missense, and synonymous variation with 90% intervals for missense and synonymous, and the three Z-scores. This is the orthogonal axis to allele frequency: a loss-of-function variant matters far more in a gene intolerant to being broken. Accepts an HGNC symbol (PCSK9) or an Ensembl gene ID (ENSG00000169174). constraint_release names the release the metrics come from: gnomAD v4.1.2 for gnomad_r4, gnomad_r3, and their subsets (gnomAD publishes no v3 constraint), gnomAD v2.1.1 for gnomad_r2_1 and its subsets, and ExAC r0.3 for exac. gnomAD publishes no subset constraint, so a subset dataset reads its parent release’s table and echoes the subset as dataset. gnomAD recommends LOEUF < 0.45 to call a gene LoF-intolerant on v4.1.2 and LOEUF < 0.35 on v2.1.1; only v4.1.2 publishes the LOEUF percentile. ExAC r0.3 publishes only pLI, the Z-scores, and observed/expected counts, so on exac the ratios, their intervals, LOEUF, and the LOEUF percentile are null, constraint_flags is empty, and pLI is the intolerance measure. A mitochondrial gene (MT-CO1, MT-TL1) has no nuclear constraint: gnomAD’s mitochondrial constraint comes back in mitochondrial_constraint instead — kind protein or rna, the same table on every dataset, exac included — and is null for every nuclear gene; the nuclear LOEUF thresholds do not apply to it. Many genes have null constraint (sparse upstream) — null fields are reported as such, never fabricated. Echoes the effective dataset and reference build. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_get_gene_constraint",
    "arguments": {
      "gene": "<gene>"
    }
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "gene": {
      "type": "string",
      "minLength": 2,
      "description": "Gene — HGNC symbol (e.g. PCSK9) or Ensembl gene ID (e.g. ENSG00000169174). Obtain a stable ID from ensembl_lookup_gene."
    },
    "dataset": {
      "description": "gnomAD dataset, default gnomad_r4: gnomad_r4 and its subset gnomad_r4_non_ukb (GRCh38); gnomad_r3 and its subsets gnomad_r3_controls_and_biobanks, gnomad_r3_non_cancer, gnomad_r3_non_neuro, gnomad_r3_non_topmed, gnomad_r3_non_v2 (GRCh38); gnomad_r2_1 and its subsets gnomad_r2_1_controls, gnomad_r2_1_non_cancer, gnomad_r2_1_non_neuro, gnomad_r2_1_non_topmed (GRCh37); exac (GRCh37). A subset keeps only part of its parent release's samples (non_cancer leaves out cancer-study samples, non_ukb UK Biobank's) and can leave a callset unchanged: gnomad_r4_non_ukb genome counts equal gnomad_r4's, and gnomad_r2_1_non_cancer genome counts equal gnomad_r2_1's. No subset publishes joint data, and gnomAD publishes no subset coverage or constraint, so those come from the parent. Echoed in output.",
      "type": "string",
      "enum": [
        "gnomad_r4",
        "gnomad_r4_non_ukb",
        "gnomad_r3",
        "gnomad_r3_controls_and_biobanks",
        "gnomad_r3_non_cancer",
        "gnomad_r3_non_neuro",
        "gnomad_r3_non_topmed",
        "gnomad_r3_non_v2",
        "gnomad_r2_1",
        "gnomad_r2_1_controls",
        "gnomad_r2_1_non_neuro",
        "gnomad_r2_1_non_cancer",
        "gnomad_r2_1_non_topmed",
        "exac"
      ]
    },
    "reference_genome": {
      "description": "Reference build. Derived from dataset when omitted: GRCh38 for gnomad_r4, gnomad_r3, and their subsets; GRCh37 for gnomad_r2_1, its subsets, and exac. If supplied it must match the dataset, or the call is rejected. Keep aligned with ensembl coordinates.",
      "type": "string",
      "enum": [
        "GRCh38",
        "GRCh37"
      ]
    }
  },
  "required": [
    "gene"
  ],
  "additionalProperties": false
}
view source ↗

gnomad_list_gene_variants

open-world

List every gnomAD variant in a gene, transcript, or region with allele frequencies, filtering allele frequencies, zygosity counts, predicted consequences with LOFTEE's loss-of-function call, and QC status, optionally filtered to one consequence class (lof, missense, synonymous, other), a LOFTEE confidence (lof_confidence HC or LC), a maximum allele frequency, and/or a maximum filtering allele frequency (faf95_max, the value to test against a disease threshold such as ACMG/AMP BA1 or BS1). Pass annotations: true to add each row's HGVS c. and p. notation, LOFTEE filter and flags, and rsIDs. Every row carries its exome and genome QC filters and a qc_status; qc_filtered counts the matching variants that failed QC in every callset carrying them (often AC0, no high-quality carrier, at AF 0), which gnomAD's browser hides by default — read them as low-confidence, not observed variants, or pass pass_only to drop them. Each call returns one page: up to limit matching variants from offset, in gnomAD's genomic order, with total counting every match; pass next_offset as offset until it is null. gnomAD lists a target only whole, so each page re-fetches the whole target — page a large gene with filters and a large limit. At offset 0, a result too large to inline is also staged on a DataCanvas table named gene_variants holding every matching variant, returned as canvas_id and table_name — call gnomad_dataframe_describe for its columns, then gnomad_dataframe_query to rank by AF, count by consequence, or group across every row rather than one page. A result that fits inline stages no table unless canvas_id is supplied, and a page past offset 0 never stages. Supply exactly one of gene, transcript_id, or region. Echoes the resolved target, the row filters, and the effective dataset and build. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_list_gene_variants",
    "arguments": {}
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "gene": {
      "description": "Gene — HGNC symbol (e.g. PCSK9) or Ensembl gene ID (e.g. ENSG00000169174). Obtain a stable ID from ensembl_lookup_gene. Mitochondrial genes (e.g. MT-TL1) are not served. Mutually exclusive with transcript_id and region; blank means omitted.",
      "anyOf": [
        {
          "type": "string",
          "minLength": 2,
          "description": "Gene — HGNC symbol (e.g. PCSK9) or Ensembl gene ID (e.g. ENSG00000169174). Obtain a stable ID from ensembl_lookup_gene."
        },
        {
          "type": "string",
          "maxLength": 0,
          "description": "Blank — the gene is treated as omitted."
        }
      ]
    },
    "transcript_id": {
      "description": "Ensembl transcript ID (e.g. ENST00000302118). Mutually exclusive with gene and region; blank means omitted.",
      "type": "string"
    },
    "region": {
      "description": "Genomic region chrom-start-stop (1-based inclusive, e.g. 1-55039447-55064852): chromosome 1–22, X, or Y with an optional chr prefix (mitochondrial regions are not served), a span (stop − start) under 2,500,000 bp, and at most ~30,000 variants. Mutually exclusive with gene and transcript_id.",
      "anyOf": [
        {
          "type": "string",
          "const": ""
        },
        {
          "type": "string",
          "pattern": "^(?:chr)?[0-9A-Z]+-\\d+-\\d+$",
          "description": "Genomic region chrom-start-stop (1-based inclusive, e.g. 1-55039447-55064852) on chromosome 1–22, X, or Y, optional chr prefix."
        }
      ]
    },
    "consequence_class": {
      "description": "Keep only variants in this gnomAD consequence category, read from each row's consequence: lof (pLoF) is transcript_ablation, splice_acceptor_variant, splice_donor_variant, stop_gained, frameshift_variant; missense (missense / inframe indel) is missense_variant, stop_lost, start_lost, inframe_insertion, inframe_deletion; synonymous is synonymous_variant; other is every other term and a null consequence. Omit to return all classes.",
      "type": "string",
      "enum": [
        "lof",
        "missense",
        "synonymous",
        "other"
      ]
    },
    "lof_confidence": {
      "description": "Keep only variants whose lof — LOFTEE's call on the row's transcript_id — equals this: HC high-confidence or LC low-confidence. A row with lof null or OS matches neither. consequence_class lof reads the consequence term instead, so it keeps LoF terms LOFTEE did not call. Omit to keep every call.",
      "type": "string",
      "enum": [
        "HC",
        "LC"
      ]
    },
    "max_af": {
      "description": "Keep only variants with allele frequency ≤ this value (0–1). Variants with null AF are always kept.",
      "type": "number",
      "minimum": 0,
      "maximum": 1
    },
    "max_faf": {
      "description": "Keep only variants whose filtering allele frequency (faf95_max) is ≤ this value (0–1) — the test against a disease threshold such as ACMG/AMP BA1 or BS1. Variants with a null faf95_max are always kept, and a notice counts them among total. gnomad_r3, its subsets, and exac list rows carry no FAF, so there it filters nothing and the notice says so.",
      "type": "number",
      "minimum": 0,
      "maximum": 1
    },
    "pass_only": {
      "description": "Keep only variants that passed gnomAD QC in at least one callset carrying them (qc_status pass), with the counts gnomAD's browser table shows by default: a row without joint data (every dataset but gnomad_r4, gnomad_r4_non_ukb included) reads ac, an, af, homozygote_count, hemizygote_count, and faf95_max from its passing callsets only, and max_af and max_faf filter on those values; gnomad_r4 rows keep their joint counts. The drop follows consequence_class, lof_confidence, max_af, and max_faf, and qc_filtered still counts the dropped rows. Omit to list every matching variant with its QC columns and a QC-failed callset's counts included — a single-position region then never reads as absent because of QC.",
      "type": "boolean"
    },
    "annotations": {
      "description": "Add five columns describing each row's transcript_id after every default column, on the rows and in the staged gene_variants table: hgvsc, hgvsp, lof_filter, lof_flags, and rsids (pipe-joined). Omitted or false, the rows and the table carry none of them. A column toggle, not a filter: it keeps every row and filters does not echo it. It widens every row by about a quarter and every page's re-fetch of the target from gnomAD by about as much.",
      "type": "boolean"
    },
    "offset": {
      "default": 0,
      "description": "Zero-based position, among the variants the filters keep, of the first one on this page. Pass next_offset from the previous call to continue.",
      "type": "integer",
      "minimum": 0,
      "maximum": 9007199254740991
    },
    "limit": {
      "default": 25,
      "description": "Matching variants on this page. Default 25; a value above 3,000 is clamped to 3,000, with a notice.",
      "type": "number",
      "minimum": 1,
      "multipleOf": 1
    },
    "canvas_id": {
      "description": "Optional canvas ID from a prior call, to reuse the same canvas. When supplied, a call at offset 0 always writes its result to the gene_variants table on that canvas, replacing (not appending to) the previous one, its columns included: the table takes this call's annotations shape — even when the result fits inline; a result with no variants removes the table. A page past offset 0 leaves the table as it was. Omit to stage on a fresh canvas only when the result is too large to inline.",
      "type": "string",
      "pattern": "^[A-Za-z0-9_-]{10}$"
    },
    "dataset": {
      "description": "gnomAD dataset, default gnomad_r4: gnomad_r4 and its subset gnomad_r4_non_ukb (GRCh38); gnomad_r3 and its subsets gnomad_r3_controls_and_biobanks, gnomad_r3_non_cancer, gnomad_r3_non_neuro, gnomad_r3_non_topmed, gnomad_r3_non_v2 (GRCh38); gnomad_r2_1 and its subsets gnomad_r2_1_controls, gnomad_r2_1_non_cancer, gnomad_r2_1_non_neuro, gnomad_r2_1_non_topmed (GRCh37); exac (GRCh37). A subset keeps only part of its parent release's samples (non_cancer leaves out cancer-study samples, non_ukb UK Biobank's) and can leave a callset unchanged: gnomad_r4_non_ukb genome counts equal gnomad_r4's, and gnomad_r2_1_non_cancer genome counts equal gnomad_r2_1's. No subset publishes joint data, and gnomAD publishes no subset coverage or constraint, so those come from the parent. Echoed in output.",
      "type": "string",
      "enum": [
        "gnomad_r4",
        "gnomad_r4_non_ukb",
        "gnomad_r3",
        "gnomad_r3_controls_and_biobanks",
        "gnomad_r3_non_cancer",
        "gnomad_r3_non_neuro",
        "gnomad_r3_non_topmed",
        "gnomad_r3_non_v2",
        "gnomad_r2_1",
        "gnomad_r2_1_controls",
        "gnomad_r2_1_non_neuro",
        "gnomad_r2_1_non_cancer",
        "gnomad_r2_1_non_topmed",
        "exac"
      ]
    },
    "reference_genome": {
      "description": "Reference build. Derived from dataset when omitted: GRCh38 for gnomad_r4, gnomad_r3, and their subsets; GRCh37 for gnomad_r2_1, its subsets, and exac. If supplied it must match the dataset, or the call is rejected. Keep aligned with ensembl coordinates.",
      "type": "string",
      "enum": [
        "GRCh38",
        "GRCh37"
      ]
    }
  },
  "required": [
    "offset",
    "limit"
  ],
  "additionalProperties": false
}
view source ↗

gnomad_get_coverage

open-world

Fetch gnomAD sequencing-coverage summary across a gene, transcript, or region — mean and median read depth, plus the mean fraction of samples covered at each depth threshold (1× through 100×), separated by exome and genome track. gnomAD reports coverage in buckets: 1 bp for a region spanning under 850 bp, wider for larger targets (bucket_size), and a gene or transcript is summarized over its exons ±75 bp, not its introns. Use this to disambiguate a true absent variant from an uncallable position: a variant missing from a well-covered region is informative, while one missing from a poorly-covered region is not. Supply exactly one of gene, transcript_id, or region. The optional coverage_source narrows to one track; by default both available tracks are returned. gnomAD publishes no coverage for subsets, so a subset dataset reads its parent release (gnomad_r4_non_ukb reads gnomad_r4) over all the parent’s samples: coverage_dataset names the release read, and a notice says so. Echoes the effective dataset and build. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_get_coverage",
    "arguments": {}
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "gene": {
      "description": "Gene — HGNC symbol (e.g. PCSK9) or Ensembl gene ID (e.g. ENSG00000169174). Obtain a stable ID from ensembl_lookup_gene. Mitochondrial genes (e.g. MT-TL1) are not served. Mutually exclusive with transcript_id and region; blank means omitted.",
      "anyOf": [
        {
          "type": "string",
          "minLength": 2,
          "description": "Gene — HGNC symbol (e.g. PCSK9) or Ensembl gene ID (e.g. ENSG00000169174). Obtain a stable ID from ensembl_lookup_gene."
        },
        {
          "type": "string",
          "maxLength": 0,
          "description": "Blank — the gene is treated as omitted."
        }
      ]
    },
    "transcript_id": {
      "description": "Ensembl transcript ID (e.g. ENST00000302118). Mutually exclusive with gene and region; blank means omitted.",
      "type": "string"
    },
    "region": {
      "description": "Genomic region chrom-start-stop (1-based inclusive, e.g. 1-55039447-55064852): chromosome 1–22, X, or Y with an optional chr prefix (mitochondrial regions are not served) and a span (stop − start) under 2,500,000 bp. Mutually exclusive with gene and transcript_id.",
      "anyOf": [
        {
          "type": "string",
          "const": ""
        },
        {
          "type": "string",
          "pattern": "^(?:chr)?[0-9A-Z]+-\\d+-\\d+$",
          "description": "Genomic region chrom-start-stop (1-based inclusive, e.g. 1-55039447-55064852) on chromosome 1–22, X, or Y, optional chr prefix."
        }
      ]
    },
    "coverage_source": {
      "description": "Restrict to one coverage track. Omit to return every available track.",
      "type": "string",
      "enum": [
        "exome",
        "genome"
      ]
    },
    "dataset": {
      "description": "gnomAD dataset, default gnomad_r4: gnomad_r4 and its subset gnomad_r4_non_ukb (GRCh38); gnomad_r3 and its subsets gnomad_r3_controls_and_biobanks, gnomad_r3_non_cancer, gnomad_r3_non_neuro, gnomad_r3_non_topmed, gnomad_r3_non_v2 (GRCh38); gnomad_r2_1 and its subsets gnomad_r2_1_controls, gnomad_r2_1_non_cancer, gnomad_r2_1_non_neuro, gnomad_r2_1_non_topmed (GRCh37); exac (GRCh37). A subset keeps only part of its parent release's samples (non_cancer leaves out cancer-study samples, non_ukb UK Biobank's) and can leave a callset unchanged: gnomad_r4_non_ukb genome counts equal gnomad_r4's, and gnomad_r2_1_non_cancer genome counts equal gnomad_r2_1's. No subset publishes joint data, and gnomAD publishes no subset coverage or constraint, so those come from the parent. Echoed in output.",
      "type": "string",
      "enum": [
        "gnomad_r4",
        "gnomad_r4_non_ukb",
        "gnomad_r3",
        "gnomad_r3_controls_and_biobanks",
        "gnomad_r3_non_cancer",
        "gnomad_r3_non_neuro",
        "gnomad_r3_non_topmed",
        "gnomad_r3_non_v2",
        "gnomad_r2_1",
        "gnomad_r2_1_controls",
        "gnomad_r2_1_non_neuro",
        "gnomad_r2_1_non_cancer",
        "gnomad_r2_1_non_topmed",
        "exac"
      ]
    },
    "reference_genome": {
      "description": "Reference build. Derived from dataset when omitted: GRCh38 for gnomad_r4, gnomad_r3, and their subsets; GRCh37 for gnomad_r2_1, its subsets, and exac. If supplied it must match the dataset, or the call is rejected. Keep aligned with ensembl coordinates.",
      "type": "string",
      "enum": [
        "GRCh38",
        "GRCh37"
      ]
    }
  },
  "additionalProperties": false
}
view source ↗

gnomad_get_variant_cooccurrence

open-world

Estimate whether two variants in the same gene sit on different haplotypes (in trans — compound heterozygous) or on the same one (in cis), from gnomAD v2.1 exome genotypes: the probability of compound heterozygosity and the genotype and haplotype counts behind it, overall and per genetic-ancestry group. For a proband with two rare heterozygous variants in a recessive gene and no parental samples. Takes exactly two GRCh37 chrom-pos-ref-alt IDs and always reads gnomad_r2_1, the only dataset gnomAD serves co-occurrence for, echoing it with GRCh37. A GRCh38 ID can name a different GRCh37 variant or none, so convert GRCh38 IDs first: gnomad_get_variant on dataset gnomad_r2_1 lists their GRCh37 IDs as candidates only when the ID fails there as variant_not_found_other_build; an ID gnomad_r2_1 finds returns a different variant. gnomAD answers only for two different coding or UTR variants of one gene, both in its exome samples with exome allele frequency of 5% or less; any other pair fails as cooccurrence_unavailable with gnomAD's reason. Each count key names variant_ids[0]'s genotype or allele first and variant_ids[1]'s second, and variant_ids can list the pair in another order than sent. The notice carries gnomAD's calibration caveat and its browser's reading of the probability. gnomAD's browser also notes when the pair is found in phase as a multi-nucleotide variant; this tool does not check for one. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_get_variant_cooccurrence",
    "arguments": {
      "variants": "<variants>"
    }
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "variants": {
      "minItems": 2,
      "maxItems": 2,
      "type": "array",
      "items": {
        "type": "string",
        "description": "GRCh37 variant ID — chrom-pos-ref-alt (1-based, e.g. 1-55505647-G-T) on chromosome 1–22, X, or Y, with an optional chr prefix."
      },
      "description": "The two variants to phase, as GRCh37 chrom-pos-ref-alt IDs. Order does not matter: variant_ids in the output gives the order the counts use."
    }
  },
  "required": [
    "variants"
  ],
  "additionalProperties": false
}
view source ↗

gnomad_search_clinvar

open-world

Search ClinVar (NCBI E-utilities) for a gene and return its classified variants — the germline clinical significance with its review status, 0–4 star rating, and conditions; any somatic oncogenicity or clinical-impact classification with its own review status and tumor types; molecular consequences, submission counts, and gnomAD-compatible identifiers (canonical SPDI, rsIDs, and a GRCh38 variant ID for gnomad_get_variant, derived from the SPDI or filled from gnomAD's own ClinVar mapping) — turning the variant-level significance gnomAD joins into a gene-panel curation view. Optionally filter by germline clinical_significance (e.g. pathogenic) and a minimum germline star rating. Each call returns one window of up to limit ClinVar records (default 10): total_found is the ClinVar count for the search terms (the match count for a mapped significance term alone, a candidate count once min_review_stars or any other term applies), and next_offset continues through the rest via offset. A window too large to inline is staged on a DataCanvas table named clinvar_variants, returned as canvas_id and table_name beside an inline preview — call gnomad_dataframe_describe for its columns, then gnomad_dataframe_query to rank or count across the window. A window that fits inline stages no table unless canvas_id is supplied. Keyless, but honors NCBI_API_KEY for a higher rate limit. Credit: ClinVar, NCBI.

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_search_clinvar",
    "arguments": {
      "gene": "<gene>"
    }
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "gene": {
      "type": "string",
      "minLength": 2,
      "description": "Gene HGNC symbol (e.g. PCSK9). ClinVar indexes HGNC symbols only — Ensembl gene IDs (ENSG…) are not resolved here, unlike the other gnomAD tools; resolve one to its symbol via ensembl_lookup_gene."
    },
    "clinical_significance": {
      "description": "Filter by ClinVar germline classification, case-insensitive, matched as whole words; underscores read as spaces. Somatic oncogenicity and clinical-impact classifications are never matched. pathogenic (Pathogenic and Likely pathogenic, combined and low-penetrance forms included), likely pathogenic, uncertain significance, vus (Uncertain significance plus VUS-high, VUS-mid, and VUS-low), likely benign, benign (Benign and Likely benign), and conflicting search ClinVar's germline classification index, so on their own their windows hold matching records only. Any other term (e.g. risk factor) narrows a text search, and its windows can return fewer rows than limit. Blank means no filter.",
      "type": "string"
    },
    "min_review_stars": {
      "description": "Keep only variants whose germline classification has at least this gold-star rating (0–4). ClinVar's review-status search also matches the review status of a somatic classification, and those records are dropped after the fetch, so a window can return fewer rows than limit.",
      "type": "integer",
      "minimum": 0,
      "maximum": 4
    },
    "offset": {
      "default": 0,
      "description": "Zero-based position of the first ClinVar record in this window. Pass next_offset from the previous call to continue.",
      "type": "integer",
      "minimum": 0,
      "maximum": 2147483647
    },
    "limit": {
      "default": 10,
      "description": "ClinVar records to fetch in this window. A mapped clinical_significance term alone narrows the search to matching records, so its windows are full; min_review_stars and any other significance term are also applied after the fetch, so their windows can return fewer rows. Default 10; a value above 1,500 is clamped to 1,500, with a notice.",
      "type": "number",
      "minimum": 1,
      "multipleOf": 1
    },
    "canvas_id": {
      "description": "Optional canvas ID from a prior call, to reuse the same canvas. When supplied, each search writes its window to the clinvar_variants table on that canvas, replacing (not appending to) the previous one — even when the window fits inline; a window with no rows removes the table. An Ensembl gene ID searches nothing and leaves the canvas as it was. Omit to stage on a fresh canvas only when the window is too large to inline.",
      "type": "string",
      "pattern": "^[A-Za-z0-9_-]{10}$"
    }
  },
  "required": [
    "gene",
    "offset",
    "limit"
  ],
  "additionalProperties": false
}
view source ↗

gnomad_dataframe_query

Run a read-only SQL SELECT against a canvas table staged by gnomad_list_gene_variants (table gene_variants) or gnomad_search_clinvar (table clinvar_variants) and return one page of the result. Use the canvas_id and table_name those tools returned to rank by allele frequency, group by consequence class, count loss-of-function variants, or filter the full set the inline preview only sampled. A page holds up to limit rows (default 25, capped at 3,000) and ends before a row that would take its row JSON past 4,000,000 characters; continue from next_offset until it is null. Each page re-runs the SQL, so stable paging needs an ORDER BY over a unique key (such as variant_id) and an unchanged table. Paging reaches the server row cap: above it total is null and later rows are reachable only by filtering or aggregating in SQL. BIGINT values, such as the staged ac and an counts and COUNT(*), arrive as JSON strings to keep their precision; CAST(col AS INTEGER) returns a number. SELECT statements only — writes, DDL, and file/HTTP table functions are rejected by the canvas gate. Call gnomad_dataframe_describe first to discover staged table and column names.

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_dataframe_query",
    "arguments": {
      "canvas_id": "<canvas_id>",
      "sql": "<sql>"
    }
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "canvas_id": {
      "type": "string",
      "pattern": "^[A-Za-z0-9_-]{10}$",
      "description": "Canvas ID returned by gnomad_list_gene_variants or gnomad_search_clinvar."
    },
    "sql": {
      "type": "string",
      "minLength": 1,
      "description": "Read-only SQL SELECT. Reference tables by the names the staging tool returned (e.g. gene_variants). Add an ORDER BY over a unique key when paging."
    },
    "offset": {
      "default": 0,
      "description": "Row offset of the page to return. Start at 0, then pass next_offset from the previous page.",
      "type": "integer",
      "minimum": 0,
      "maximum": 9007199254740991
    },
    "limit": {
      "default": 25,
      "description": "Maximum rows on the page. A page also ends before a row that would take its row JSON past 4,000,000 characters. Default 25; a value above 3,000 is clamped to 3,000, with a notice.",
      "type": "number",
      "minimum": 1,
      "multipleOf": 1
    }
  },
  "required": [
    "canvas_id",
    "sql",
    "offset",
    "limit"
  ],
  "additionalProperties": false
}
view source ↗

gnomad_dataframe_describe

List the tables staged on a canvas and their columns (name and type) so you can write correct SQL for gnomad_dataframe_query. Use the canvas_id returned by gnomad_list_gene_variants or gnomad_search_clinvar. Returns one entry per table with its row count and column schema.

read
invocation
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "gnomad_dataframe_describe",
    "arguments": {
      "canvas_id": "<canvas_id>"
    }
  }
}
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "canvas_id": {
      "type": "string",
      "pattern": "^[A-Za-z0-9_-]{10}$",
      "description": "Canvas ID returned by a prior staging call (gnomad_list_gene_variants or gnomad_search_clinvar)."
    }
  },
  "required": [
    "canvas_id"
  ],
  "additionalProperties": false
}
view source ↗

disabled 1

gnomad_dataframe_drop

Drop a named table from a canvas to reclaim memory. Opt-in and off by default — a deliberate mutation on the otherwise read-only research surface. Use the canvas_id and table_name from a prior staging call. Per-table TTL already reclaims memory automatically, so this is a convenience for operators who want manual control.

disabledwould be destructive

Disabled. Canvas table drop is turned off in this deployment — the research surface stays read-only.

GNOMAD_DATAFRAME_DROP_ENABLED=true
schema
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "canvas_id": {
      "type": "string",
      "pattern": "^[A-Za-z0-9_-]{10}$",
      "description": "Canvas ID holding the table to drop."
    },
    "table_name": {
      "type": "string",
      "minLength": 1,
      "description": "Name of the table to drop (e.g. gene_variants, clinvar_variants)."
    }
  },
  "required": [
    "canvas_id",
    "table_name"
  ],
  "additionalProperties": false
}
view source ↗

Resources

2

Population record for one gnomAD variant — AC/AN/AF overall and per ancestry, the filtering allele frequency (faf95), counts, flags, consequence with its HGVS and LOFTEE call, LoF curations, in-silico predictors, and joined ClinVar significance. Mirrors gnomad_get_variant. The dataset segment (e.g. gnomad_r4, or a subset such as gnomad_r4_non_ukb) makes the URI self-describing. On a gnomad_r4 or gnomad_r3 subset, faf95 and each callset's faf95 are null: gnomAD's record there carries the parent release's FAF, not the subset's, and a resource carries no notice to say so — read the parent's segment for its faf95. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

uri gnomad://variant/{dataset}/{variantId} mime application/json

gnomAD loss-of-function constraint for a gene — pLI, LOEUF (oe_lof_upper) with CI and gnomAD’s LOEUF percentile (v4.1.2 only), observed/expected ratios with 90% intervals for missense and synonymous, and Z-scores, with constraint_release naming the release they come from. Mirrors gnomad_get_gene_constraint: the exac segment serves ExAC r0.3 constraint (pLI, Z-scores, and counts only), and gnomad_r3 serves the GRCh38 gnomAD v4.1.2 table because gnomAD publishes no v3 constraint. gnomAD publishes no subset constraint either, so a subset segment reads its parent release’s table and the record echoes the subset as dataset. gnomAD recommends LOEUF < 0.45 to call a gene LoF-intolerant on v4.1.2 and LOEUF < 0.35 on v2.1.1; on exac, which has no LOEUF, pLI (>0.9 intolerant) is the intolerance measure. A mitochondrial gene’s constraint comes back in mitochondrial_constraint (kind protein or rna, sums of maximum heteroplasmy rather than variant counts), the same table on every segment and null for a nuclear gene. The gene segment is an HGNC symbol or Ensembl gene ID. Data source: gnomAD (Broad Institute) — https://gnomad.broadinstitute.org/

uri gnomad://gene/{dataset}/{gene}/constraint mime application/json

Prompts

1

Guided rare-disease variant-triage workflow over gnomAD: pull the variant population record, weigh it against gene loss-of-function constraint, and — critically — confirm the position is well-covered before concluding a variant is absent. Emits the exact tool chain in order.

  • variantrequired — Variant to triage — a chrom-pos-ref-alt variantId (e.g. 1-55051215-G-GA) or an rsID (rs11591147).
  • gene — Gene symbol or Ensembl ID for the constraint step (e.g. PCSK9). Omit (or leave blank) if not yet known.
  • dataset — gnomAD dataset to use: gnomad_r4 and its subset gnomad_r4_non_ukb (GRCh38); gnomad_r3 and its subsets gnomad_r3_controls_and_biobanks, gnomad_r3_non_cancer, gnomad_r3_non_neuro, gnomad_r3_non_topmed, gnomad_r3_non_v2 (GRCh38); gnomad_r2_1 and its subsets gnomad_r2_1_controls, gnomad_r2_1_non_cancer, gnomad_r2_1_non_neuro, gnomad_r2_1_non_topmed (GRCh37); exac (GRCh37). A subset reads its parent release's coverage and constraint. faf95 is null on a gnomad_r4 or gnomad_r3 subset; look the variant up on the parent. Omit or leave blank to use the server default.

Extensions

1

io.modelcontextprotocol/skills

{
  "directoryRead": true
}