bio-clinical-databases-gnomad-frequencies
Queries gnomAD v4 (807k samples), v3, v2.1.1, and constraint metrics with grpmax FAF95, bottleneck-group exclusion, LOEUF interpretation, SV/CNV/mtDNA catalogs, and Whiffin max-credible-AF framework. Use when filtering rare variants, applying ACMG BS1/BA1, ranking genes by LoF intolerance, or selecting between v2 (GRCh37 + chrX/Y constraint) and v4 (GRCh38 + 807k samples).
npx skills add BioTender-max/awesome-bio-agent-skills --skill gnomad-frequencies --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Describes a Python-based skill that queries gnomAD frequency data across v4, v3, and v2.1.1, including FAF95 grpmax, LOEUF, SV/CNV/mtDNA catalogs, and Whiffin max-credible-AF framework. It supports ACMG BS1/BA1 assessment guidance, gene-level constraint ranking, and dataset-version decision logic for v2 vs v4 use cases.
How it works
- Provides GraphQL-based single-variant queries to retrieve exome/genome AF, grpmax FAF95, and per-ancestry breakdown via the gnomAD GraphQL API at https://gnomad.broadinstitute.org/api, with dataset options including gnomad_r4, gnomad_r3, and gnomad_r2_1.
- Exposes a helper grpmax_faf95(payload) to extract the ACMG-grade FAF95 from exome/genome responses, prioritizing FAF95 fields under the appropriate dataset.
- Includes functions to apply ACMG BS1/BA1 using the max-credible-AF formula (Whiffin 2017) and a comparison against grpmax FAF95, with ba1_threshold defaulting to 0.05.
- Offers a gene-constraint retrieval function query_gene_constraint(gene_symbol, dataset) that fetches LOEUF-related metrics and notes chrX/Y constraint limitations (v4 lacks X/Y constraint).
- Provides a bulk-variant filtering example using Hail on Google Cloud Storage (GCS) with a v4 exomes table, demonstrating how to filter by grpmax FAF95 thresholds.
When to use it
- Use for ACMG BS1/BA1 assessments and to determine allele frequency thresholds in a conservative decision framework.
- Use for gene-level LoF intolerance ranking using LOEUF metrics from the appropriate v4 (autosomes) or v2.1.1 (chrX/Y) datasets.
- Use for dataset-version decisions between v2/v3/v4 depending on GRCh37/GRCh38, sample counts, and the presence of chrX/Y constraint data.
- Use for bulk rare-variant filtering at cohort scale when API rate limits are prohibitive.
What it can touch
- GraphQL API at https://gnomad.broadinstitute.org/api
- Python modules: requests
- Hail Tables on Google Cloud Storage at gs://gcp-public-data--gnomad/
- The code examples reference dataset identifiers like gnomad_r4, gnomad_r3, gnomad_r2_1, and files under gs://gcp-public-data--gnomad/release/4.1/ht/exomes/gnomad.exomes.v4.1.sites.ht
Caveats
- v4 constraint metrics exist for autosomes; chrX and chrY constraints are not released in v4, so the code suggests fallback to v2.1.1 for X/Y in some cases.
- The RP examples emphasize using grpmax FAF95 rather than raw AF for BS1/BA1; the BA1 threshold is a default 5% but may be overridden by VCEP guidance.
- The graphQL example notes dataset options and emphasizes that liftover and assembly differences can affect variant representation across versions.
## Version Compatibility Reference examples tested with: requests 2.31+, hail 0.2.130+, pandas 2.2+, myvariant 1.0+. Current gnomAD release is **v4.1 (May 2024)**; v4.1 fixed the v4.0 AN under-counting issue that inflated rare-variant AF estimates by 5-10%. Before using code patterns, verify installed versions match. If versions differ: - Python: `pip show <package>` then `help(module.function)` to check signatures - Hail: `hl.version()`; pin to >=0.2.130 for v4 schema If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying. The gnomAD browser GraphQL API at `https://gnomad.broadinstitute.org/api` is the supported public endpoint; Hail Tables on Google Cloud Storage at `gs://gcp-public-data--gnomad/` are the supported bulk access. # gnomAD Frequency Queries and Constraint **'How rare is this variant in the general population?'** -> Pull allele frequency, grpmax FAF95 (the ACMG-grade frequency), LOEUF gene-level constraint, structural variant catalog, mtDNA frequencies, and the appropriate dataset version per use case. - Python (single variant): GraphQL via `requests.post('https://g
- Version Compatibility
- v2.1.1 / v3.1.2 / v4.x: When to Use Which
- v4 Ancestry Groups: popmax -> grpmax Terminology
- Filtering Allele Frequency (FAF95): The ACMG-Grade AF
- Constraint Metrics: pLI, LOEUF, missense Z
- Subsets: noncancer, nonneuro, controls
- SV Catalog and CNV
- mtDNA (Laricchia 2022 Genome Res 32:569)
- VEP Version Pinning
- Decision Tree by Query Scenario
- Single Variant Query (GraphQL)
- ACMG BS1/BA1 Application
- Gene-Level Constraint (LOEUF)
- Bulk Query via Hail (cohort-scale)
What does the bio-clinical-databases-gnomad-frequencies skill do?
Queries gnomAD v4 (807k samples), v3, v2.1.1, and constraint metrics with grpmax FAF95, bottleneck-group exclusion, LOEUF interpretation, SV/CNV/mtDNA catalogs, and Whiffin max-credible-AF framework. Use when filtering rare variants, applying ACMG BS1/BA1, ranking genes by LoF intolerance, or selecting between v2 (GRCh37 + chrX/Y constraint) and v4 (GRCh38 + 807k samples).
How do I install it?
Run `npx skills add BioTender-max/awesome-bio-agent-skills --skill gnomad-frequencies --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From BioTender-max/awesome-bio-agent-skills, a repository with 135 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.