The Alliance for Genomic Discovery
A consortium NashBio launched with Illumina and ten pharmaceutical partners. 312,000+ whole genomes sequenced from the BioVU® biobank, each one linked to a longitudinal clinical record.
- 312,000+
- whole genomes sequenced
- 10
- pharmaceutical partners
- 2 years
- from first genome to completion
Genetic evidence changes the odds
A 2024 Nature study found that drug programs with genetic evidence behind them succeed 2.6 times more often than those without. The genomic resources large enough to supply that evidence have mostly been built on homogeneous Western European ancestry, and on questionnaire data rather than clinical records, which limits what they can say about how a variant behaves in a real patient population.
Built for discovery
AGD launched in January 2022, reached 250,000 samples in March 2025, and now stands at 312,000+ whole genomes, making it one of the fastest large clinical genomic datasets assembled to date. De-identified DNA from NashBio's VUMC BioVU® biobank was sequenced and integrated with EHR-derived phenotypes to form a shared platform for the member companies.
The build prioritized ancestral diversity and linked every sample to longitudinal clinical phenotypes, so findings hold up across the populations a therapy will actually treat.
Members
How AGD was built
- January 2022Launched by NashBio and Illumina to generate a powerful new genomics resource
- January 2023Data generation begins in collaboration with deCODE genetics
- March 2025250,000 whole genome milestone reached (currently at 312,000+)
- Next phaseOngoing sequencing and AGD 2.0 will drive the resource to 365,000 whole genomes and 50,000 proteomes
| Launched | January 2022, by NashBio and Illumina |
|---|---|
| Members | Ten leading pharmaceutical partners using genomics to advance R&D |
| Scale | 250,000 whole genomes, milestone reached March 2025 currently at 312,000+ |
| Sample source | De-identified DNA from NashBio's VUMC BioVU® biobank |
| Sequencing | deCODE genetics, an Amgen subsidiary |
| Analysis | Illumina's DRAGEN pipeline and Connected Analytics |
| Linkage | Every sample tied to EHR-derived longitudinal phenotypes |
The AGD dataset is unique for its diversity and depth and has already proven its value by empowering us to make important discoveries we would not have made without it.
What comes next
AGD's next phase continues to add whole genomes and new multiomic layers, starting with proteomics, to probe disease mechanisms and surface new intervention targets.
Membership inquiries and questions about access go straight to our team.
NashBio data is also available outside AGD. See Data products for what your team can license directly.
Frequently Asked Questions
- What is the Alliance for Genomic Discovery?
- The Alliance for Genomic Discovery is a consortium launched in January 2022 by NashBio and Illumina with ten pharmaceutical partners. It sequenced 312,000+ whole genomes from NashBio's VUMC BioVU® biobank, each linked to a longitudinal electronic health record, to give members a shared resource for drug target discovery and development.
- Which companies are AGD members?
- AGD members are AbbVie, Alnylam, Amgen, AstraZeneca, Bayer, Bristol Myers Squibb, GSK, Merck, Novo and the Regeneron Genetics Center. Members share access to the whole-genome and linked clinical data for target discovery, validation and therapeutic research.
- How many genomes does AGD include?
- AGD reached 250,000 whole genomes in March 2025, roughly two years after data generation began in January 2023, and has since grown to 312,000+. Sequencing was carried out by deCODE genetics through Illumina's DRAGEN pipeline, reaching 100,000 genomes in under two years against the three to five years earlier efforts took.
- Can I access AGD data without being a member?
- AGD data is available to member companies. NashBio licenses whole genome sequencing and linked clinical data outside the consortium as a standard data product, with 312K+ sequences at 30X coverage delivered as VCF or gVCF alongside the structured clinical foundation.