Solutions
The genomics API for developers.
Ingest VCF, WGS, 23andMe, and AncestryDNA exports through a single endpoint. Get back ClinVar-annotated variants, polygenic risk scores, pharmacogenomic guidance, and ancestry insights — over REST and MCP.
What the genomics API handles
Variant annotation
Every variant is annotated against ClinVar for clinical significance and CPIC for pharmacogenomic guidance (levels A and B). Homozygous, heterozygous, and compound-heterozygous calls are resolved automatically.
Polygenic risk scores
Risk scores are computed from GWAS catalog data for cardiovascular, metabolic, neurodegenerative, and cancer phenotypes. Every score includes the population reference and confidence interval.
Pharmacogenomics
CPIC level A and B gene-drug pairs are extracted and translated into plain-language guidance. Your users see which medications may need dose adjustment — with citations back to CPIC guidelines.
Ancestry and heritage
Population-genetics analysis returns regional ancestry proportions, migration-pathway inference, and heritage composition. Render it as a map, a donut chart, or a table — the data is yours.
Aging hallmarks mapping
Every relevant variant is mapped to López-Otín's hallmarks of aging: genomic instability, telomere attrition, epigenetic alterations, and more. Useful for longevity products and healthspan tracking.
Idempotent ingestion
Upload the same VCF twice and get the same job ID back. No duplicate processing, no double billing. WGS jobs go through an async queue with status polling.
How the pipeline works
From raw genetic data to structured observations in four steps. Every step runs identically self-hosted or on the hosted API.
Upload
POST a VCF, VCF.GZ, 23andMe, or AncestryDNA export. Idempotency keys prevent duplicate processing.
Annotate
The pipeline annotates every variant against ClinVar and CPIC, computes polygenic scores, and resolves compound-heterozygous calls.
Analyze
Run gene-level or phenome-wide analysis on demand. Every result includes citations and population references.
Render
Get dashboard specs your UI can render directly, or query individual observations over REST and MCP.
Combine with
Wearable Data API →
Pull WHOOP, Oura, Apple Health, and Garmin data through one interface. Combine wearable metrics with genomics for a complete picture.
Biomarker API →
Send a CSV or JSON lab panel. Get back normalized biomarkers, domain scores, derived metrics, longitudinal trends, and retest reminders. Correlate lab results with genetics and wearables.
Questions about the genomics API
What file formats does the genomics API accept?
VCF and VCF.GZ including full WGS through an async queue, plus 23andMe and AncestryDNA exports. All processing runs idempotently so duplicate uploads are safe.
Does it annotate variants?
Yes. Every variant is annotated against ClinVar for clinical significance and CPIC for pharmacogenomic guidance (levels A and B). Polygenic risk scores are computed from GWAS catalog data and mapped to aging hallmarks.
How long does WGS processing take?
Compact WGS jobs (30x) typically complete in 5–20 minutes through the async queue. Deeper coverage or multi-sample jobs take proportionally longer. The API returns a job ID you can poll.
Can I use this for ancestry and heritage reports?
Yes. The pipeline includes population-genetics analysis: regional ancestry proportions, migration-pathway inference, and heritage composition you can surface in any UI.
Is the genomics pipeline open source?
Yes. The full pipeline is on GitHub and runs identically self-hosted or on the hosted API. The hosted version adds EU infrastructure, queue management, and no-ops maintenance.
Add genomics to your product without building a bioinformatics team.