WorkflowHub: a registry for computational workflows

Area of Science:

• Computational biology
• Bioinformatics
• Scientific computing

Background:

• Computational workflows are increasingly vital for scalable, reproducible data analysis.
• Sharing workflows facilitates knowledge transfer, reduces redundant effort, and promotes best practices.
• Current workflow discoverability is hindered by diverse engines and lack of integration into research practices.

Purpose of the Study:

• To introduce WorkflowHub as a unified registry for computational workflows.
• To enhance the findability, accessibility, interoperability, and reusability (FAIR) of workflows.
• To promote workflows as citable scholarly artifacts and support their lifecycle.

Main Methods:

• WorkflowHub acts as a central registry, linking to community repositories.
• It supports the entire workflow lifecycle, from creation to execution and sharing.
• Interoperability with diverse platforms, services, and external registries is a key feature.

Main Results:

• WorkflowHub provides a unified platform for discovering and accessing computational workflows.
• It enhances the FAIRness of workflows by standardizing metadata and accessibility.
• The registry supports credit assignment and promotes workflows as scholarly outputs.
• Global reach with hundreds of research organizations and over 800 registered workflows.

Conclusions:

• WorkflowHub addresses the challenge of scattered and difficult-to-find computational workflows.
• It fosters a more collaborative and efficient research environment by promoting workflow sharing and reuse.
• By enhancing FAIR principles, WorkflowHub advances scientific reproducibility and productivity.