Dockstore

Dockstore is a community-driven platform for publishing, discovering and launching containerized bioinformatics tools and multi-step workflows. It combines Docker images with standardized workflow descriptors (Common Workflow Language, WDL, Nextflow, Galaxy and others) so workflows become machine-readable, versioned, and portable. Founded to promote FAIR practices in computational biology, Dockstore serves as an entry point for researchers to share reusable analytical recipes, attach rich metadata and documentation, and expose them to automated execution and third‑party services. At its core Dockstore indexes workflow descriptors and the containers they reference, and it implements the Global Alliance for Genomics and Health (GA4GH) Tool Registry Service (TRS) API. That makes published entries discoverable by other tools and platforms and enables automated export and launching. Entries can be registered in multiple ways: Dockstore can index descriptors directly from source control (GitHub, Bitbucket), you can host descriptors on Dockstore itself for quick prototyping, or use a hybrid approach where source repositories, container registries (Quay.io, Docker Hub) and Dockstore are wired together. The platform recognizes versions and snapshots, supports test parameter files and “checker” workflows for validation, and can generate DOIs (via Zenodo) for tagged releases to improve citation and reproducibility. Capabilities include a web UI for browsing, organization pages for labs or projects, and a CLI plus webservice components for automation and integration. Dockstore parses metadata, displays execution and validation metrics where available, and flags platform verifications so users know which workflow versions have been run successfully on partner environments. It exposes a Swagger/OpenAPI surface for programmatic use and is integrated with launch partners and cloud platforms: entries can be exported or launched into Terra, AnVIL, Elwazi and BioData Catalyst, among others. The platform also supports emerging descriptor types (e.g., Snakemake and R Jupyter notebooks in preview), and encourages authors to provide test data and instructions to make workflows runnable across different compute backends. Common use-cases include sharing a validated pipeline for standardized analyses (for example, viral genome assembly and variant calling pipelines used during the COVID-19 response), hosting training or teaching workflows, and curating collections of community best-practice workflows (the Dockstore catalog includes contributions from consortia such as Broad Institute projects and UCSC). Researchers use Dockstore to find a workflow, inspect its descriptors and container version, run it locally via the Dockstore CLI, or launch it on a supported cloud platform without re-writing descriptors. Project owners use it to track usage metrics, attach DOIs for reproducible references in publications, and maintain multiple versions and verification metadata for each workflow. For developers and integrators, Dockstore provides a modular stack: the webservice that serves the API and indexing functionality, a separate UI repo for the front end, and a CLI client for command-line and automated workflows. The project is open source and includes guidance for local development, CI integration, and testing; the webservice exposes a Swagger UI for exploring endpoints. Dockstore’s adherence to standards (TRS, standardized descriptors) lowers the friction of plugging in new execution backends and cloud launchers, making it straightforward to add a new partner or to build custom automation that queries or imports registry entries. Dockstore is valuable where reproducibility, discoverability and portability matter. By combining containerized software, machine-readable descriptors, and standards-based APIs, it helps labs, consortia and platforms publish workflows that others can find, verify and run with minimal friction. Whether you’re publishing a single-tool Docker image with a CWL wrapper or maintaining an organization’s library of complex multi-stage pipelines, Dockstore provides the metadata, versioning, validation hooks and platform integrations needed to make bioinformatics analyses reusable and portable.