BioVault: A privacy-first data visitation platform for equitable global collaboration in biomedicine

Abstract

Biomedical datasets representing diverse populations are essential for advancing precision medicine, yet remain siloed due to regulatory, sovereignty, and privacy constraints. Existing data-sharing solutions remain limited. Centralized repositories and Trusted Research Environments (TREs) require data migration into external infrastructure, introducing governance complexities, cost barriers, and constraints on accountable reuse. Secure computation frameworks from modern cryptography, and federated learning models demand deep domain expertise and substantial engineering effort, limiting their accessibility in resource-constrained environments. These barriers disproportionately impact under-resourced institutions, limiting equitable participation in global collaborations. Here, we introduce BioVault (https://www.biovault.net), an open-source platform for privacy-first biomedical collaboration through a peer-to-peer data visitation network, where analysis code travels to data rather than data being transferred to centralized systems. BioVault can be deployed as a desktop application or command-line interface, enabling out-of-the-box use for participants in diverse resource settings. BioVault supports both clinical and research workflows, and with built-in permissioning, audit trails, and local governance controls, it enables data holders and collaborators to retain oversight and control while participating in collaborative research. We demonstrate BioVaults utility in enabling two cross-border collaborations with under-resourced communities: a genome-wide association study of Type 2 Diabetes in Circassian and Chechen populations in Jordan, and allele frequency estimation across Caribbean cohorts. In both cases, analyses were executed locally without data export. We further demonstrate compatibility for user-defined secure, federated computation protocols built on state-of-the-art cryptographic tools (multiparty computation and homomorphic encryption), by lowering barriers to their deployment. Together, these results establish BioVault as a general-purpose framework for decentralized analytics that reduces the technical and financial constraints to biomedical collaboration - enabling diverse institutions to participate as equal partners in global discovery, while preserving privacy and data sovereignty.