[Integrations](https://spiral.ai/integrations) › [Instinct](https://spiral.ai/integrations/instinct) › [gnomAD gene variation search](https://spiral.ai/integrations/instinct/gnomad-gene-variation-search)

# How to Use gnomAD gene variation search in Instinct

Returns observed DNA changes in a gene with population frequencies and consequence annotations. Use it to build a shortlist of rare, missense, or predicted loss-of-function alleles before inspecting individual sites.

|  |  |
| --- | --- |
| Model | gnomAD gene variation search |
| Available through | Alexandria |
| Category | Health |
| Price | $0.00873 per call. |
| Returns | json |
| Runs | In the background |
| Setup | About 2 minutes, with a Spiral API key |

## Prompts to try with gnomAD gene variation search in Instinct

- Every month, use gnomAD gene variation search for SCN2A to retrieve high-confidence predicted loss-of-function alleles below a frequency of 0.001 and report changes from the previous run.
- Every quarter, use gnomAD gene variation search for CFTR to retrieve missense alleles between frequencies of 0.0001 and 0.001 and compare the results with the previous run.
- Every quarter, use gnomAD gene variation search for PCSK9 to review high-confidence predicted loss-of-function alleles and summarize changes in their returned frequencies and annotations.
- Every Monday at 9am: use gnomAD gene variation search to look this up and give me the result as a table.
- Pull the latest numbers from gnomAD gene variation search and tell me what changed since last week.

## How to set up gnomAD gene variation search in Instinct

Pick the way that suits you. Each takes about two minutes.

### Use the Spiral skill

1. **Send Instinct this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use Spiral's MCP server

1. **Create an API key**: In the console, open API keys and create a key for Instinct. Copy it.

2. **Send Instinct this message**: Instinct reads Spiral's skill and asks for your API key.

```text
set up https://spiral.ai/SKILL.md
```

3. **Give Instinct the key**: When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run gnomAD gene variation search**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "alexandria",
    "model": "gnomad-broadinstitute-org/variants/search",
    "input": {
      "gene": "<gene>"
    }
  }'
```

Then paste one of the prompts above: Instinct finds gnomAD gene variation search on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/instinct

## gnomAD gene variation search tools and prices

- [Population allele frequencies and gene constraint (Broad Institute) · Search](https://spiral.ai/tools/alexandria/gnomad-broadinstitute-org/variants/search) (Alexandria): $0.00873 per call.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| sort | position, af_desc | position | position (genomic order) or af_desc (most common first). |
| limit | 1 to 100 | 25 | Variants per page. |
| max_af | 0 to 1 | — | Keep variants with at most this allele frequency. |
| min_af | 0 to 1 | — | Keep variants with at least this allele frequency. |
| offset | 0 to 150000 | 0 | Matching variants to skip. |
| dataset | gnomad_r4, gnomad_r3, gnomad_r2_1 | — | gnomad_r4 (v4.1, 730,947 exomes and 76,215 genomes, GRCh38; the default), gnomad_r3 (v3.1.2, 76,156 genomes, GRCh38) or gnomad_r2_1 (v2.1.1, 125,748 exomes and 15,708 genomes, GRCh37). Defaults to the dataset of a pasted browser URL, else gnomad_r4. |

## What can you do with gnomAD gene variation search in Instinct?

Maintain a gnomAD research digest for a fixed set of genes and public reference alleles. Compare returned counts and annotations with the previous run, including changes that follow a gnomAD release update.

### Build a rare-allele research shortlist

Survey CFTR for rare coding changes, then inspect individual sites for population frequencies and homozygote counts before selecting candidates for further research.

### Review predicted loss-of-function evidence

Find predicted loss-of-function alleles in PCSK9 and examine their population counts. Keep the confidence annotations beside the frequency evidence when preparing a target research brief.

### Check ancestry-specific enrichment

Start with observed changes in HBB, then compare genetic ancestry frequencies for selected alleles. Identify cases where an overall frequency hides a much higher frequency in one ancestry group.

### Prepare a functional assay panel

Find missense alleles in BRCA1 within a chosen frequency range, then inspect population and homozygote counts for the proposed assay candidates.

### Refresh a gene evidence report

Rerun a SCN2A search after a gnomAD release change and revisit selected allele frequencies. Report changes in the research shortlist separately from changes in population counts.

### Investigate conflicting population evidence

Review a gene's observed alleles, then inspect selected sites for differences between exome and genome frequencies. Highlight discordance before combining the evidence in a research report.

## Overview

- [Alexandria](https://spiral.ai/providers/alexandria)
- [All Instinct guides](https://spiral.ai/integrations/instinct)

## gnomAD gene variation search in other agents

- [Amp](https://spiral.ai/integrations/amp/gnomad-gene-variation-search)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/gnomad-gene-variation-search)
- [Claude](https://spiral.ai/integrations/claude-desktop/gnomad-gene-variation-search)
- [Claude Code](https://spiral.ai/integrations/claude-code/gnomad-gene-variation-search)
- [Cline](https://spiral.ai/integrations/cline/gnomad-gene-variation-search)
- [Codex](https://spiral.ai/integrations/codex/gnomad-gene-variation-search)
- [Cursor](https://spiral.ai/integrations/cursor/gnomad-gene-variation-search)
- [Devin](https://spiral.ai/integrations/devin/gnomad-gene-variation-search)
- [dots](https://spiral.ai/integrations/dots/gnomad-gene-variation-search)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/gnomad-gene-variation-search)
- [Gemini](https://spiral.ai/integrations/gemini/gnomad-gene-variation-search)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/gnomad-gene-variation-search)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/gnomad-gene-variation-search)
- [goose](https://spiral.ai/integrations/goose/gnomad-gene-variation-search)
- [Grok](https://spiral.ai/integrations/grok/gnomad-gene-variation-search)
- [Grok Build](https://spiral.ai/integrations/grok-build/gnomad-gene-variation-search)
- [Hermes](https://spiral.ai/integrations/hermes/gnomad-gene-variation-search)
- [Manus](https://spiral.ai/integrations/manus/gnomad-gene-variation-search)
- [Muse](https://spiral.ai/integrations/muse/gnomad-gene-variation-search)
- [Muse Code](https://spiral.ai/integrations/muse-code/gnomad-gene-variation-search)
- [n8n](https://spiral.ai/integrations/n8n/gnomad-gene-variation-search)
- [Notion](https://spiral.ai/integrations/notion/gnomad-gene-variation-search)
- [OpenClaw](https://spiral.ai/integrations/openclaw/gnomad-gene-variation-search)
- [opencode](https://spiral.ai/integrations/opencode/gnomad-gene-variation-search)
- [Perplexity](https://spiral.ai/integrations/perplexity/gnomad-gene-variation-search)
- [Raycast](https://spiral.ai/integrations/raycast/gnomad-gene-variation-search)
- [VS Code](https://spiral.ai/integrations/vscode/gnomad-gene-variation-search)
- [Windsurf](https://spiral.ai/integrations/windsurf/gnomad-gene-variation-search)
- [Zapier](https://spiral.ai/integrations/zapier/gnomad-gene-variation-search)

## More Instinct guides

- [gnomAD gene constraint](https://spiral.ai/integrations/instinct/gnomad-gene-constraint)
- [851 Labs Background Remover](https://spiral.ai/integrations/instinct/851-labs-background-remover)
- [gnomAD allele frequencies](https://spiral.ai/integrations/instinct/gnomad-allele-frequencies)
- [Age Modify](https://spiral.ai/integrations/instinct/age-modify)
- [dbSNP population allele frequencies](https://spiral.ai/integrations/instinct/dbsnp-population-allele-frequencies)
- [Age Progression](https://spiral.ai/integrations/instinct/age-progression)

## Questions about gnomAD gene variation search in Instinct

**Can Instinct use gnomAD gene variation search?**

Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run gnomAD gene variation search like any built-in tool.

**How much does gnomAD gene variation search cost in Instinct?**

$0.00873 per call. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.

**Do I need an Alexandria account to use gnomAD gene variation search in Instinct?**

You need no Alexandria account. Instinct uses one Spiral API key for every tool, and Spiral runs gnomAD gene variation search on its own Alexandria account.

**How long does it take to set up gnomAD gene variation search in Instinct?**

About two minutes: create an API key, then send Instinct this message, then give Instinct the key.

**Does Instinct wait while gnomAD gene variation search runs?**

gnomAD gene variation search runs in the background. Instinct starts the run, checks on it and gives you the json when it's ready.

**How large is the gnomAD reference?**

The v4 reference contains 730,947 exomes and 76,215 genomes on GRCh38. These are aggregate research data from contributing sequencing projects.

**Are the ancestry frequencies representative of everyone?**

gnomAD reports frequencies by inferred genetic ancestry group, but representation is uneven. The proportion of European ancestry samples is higher in v4 than in earlier releases, partly because of UK Biobank inclusion.

**Can annotations change between gnomAD releases?**

Yes. In v4.1.1, gnomAD changed LOFTEE's GERP distance threshold to 0.0 and reannotated predicted loss-of-function confidence, so a release change can alter a shortlist even without a new population count.

**Does a missing count mean an allele was never observed?**

A No data flag means no defined genotypes were available for that sequencing data type. AC0 has a different meaning: the allele count became zero after low-quality genotypes were removed.

**Why can exome and genome frequencies disagree?**

gnomAD explicitly tests for frequency differences between those datasets. Its v4.1 discordance warning uses a contingency-table or CMH test p-value below 0.0001.

