Compute on Encrypted Genomic Data

A platform for privacy-preserving genomic analysis using FHE and federated computing.

git clone https://github.com/barisozmen/securegenomics.git && cd securegenomics && bash setup.sh

Get started paste this in terminal, or inspect the GitHub repo and setup.sh script

The super simple workflow for secure genomic analysis

See pricing ->

$ securegenomics login

Choose an open-source protocol

$ securegenomics keygen <project-id>

$ securegenomics upload <project-id> data.vcf

$ securegenomics run <project-id>

$ securegenomics result <project-id> (decrypt locally)

Built for biobanks

Monetize datasets and collaborate with researchers without compromising contributor privacy.

Built for researchers

Run allele-frequency studies, GWAS, and cryptographically verifiable science across institutional silos.

Zero-trust compute

The server stores public context, ciphertext, encrypted results, and metadata only. It never holds private crypto context, private keys, or plaintext genomic data.

Security through transparency

Open-source CLI

The security model does not depend on hidden client code. The CLI, protocol shape, and submitted public context can be inspected, reproduced, and challenged by the community.

Open source code ->

Kerckhoffs's principle

Gencrypt is designed so privacy depends on private keys and local cryptographic context, not secret algorithms. Protocol code and server behavior can be public while the server still sees only public context, ciphertext, encrypted results, and metadata.

Case studies

Three research paths from the open SecureGenomics repo: multi-party genomic studies, a protocol-oriented path for privacy-preserving GWAS, and sensitive allele-frequency analysis.

population-scale studies

Many encrypted genomes, one aggregate result

The multi-player demo shows independent data owners contributing encrypted genomes to the same researcher project before the server runs the FHE computation on ciphertext.

securegenomics upload <project-id> data.vcf

securegenomics view <project-id>

securegenomics run <project-id>

Multi-player demo ->

GWAS

Association studies without plaintext pooling

The README names GWAS as an intended analysis class. The CLI guide supplies the protocol-oriented workflow for local keys, encrypted uploads, and ciphertext computation across silos.

securegenomics protocol list

securegenomics keygen <project-id>

securegenomics run <project-id>

CLI guide ->

allele-frequency

The first concrete protocol wedge

The README points to an open sensitive allele-frequency protocol: encode genotype dosages locally, encrypt with the project public context, and aggregate frequencies over ciphertext.

encode.py

encrypt.py

circuit.py

Allele-frequency protocol ->