Technical foundation

The open-source framework the Data Intuitive platform runs on. Viash packages research-grade tools into versioned, containerised components and assembles them into production-grade omics pipelines.
Keywords

Viash, Open-source bioinformatics framework, Versioned components, Containerised pipelines, Nextflow integration, Production-grade pipelines, Reproducible bioinformatics, Docker, OpenPipeline, Data Intuitive, Workflow components, Argument parser, Unit tests, Multi-language scripts

Back to the platform Technical foundation

From research-grade tools to production-grade pipelines

We author Viash, the open-source framework the workflows run on. It packages research-grade tools into versioned, containerised components, then assembles them into production-grade omics pipelines. Five-plus years in production, with named pharma and research institutes on board. The foundation is open-source and community-maintained, so you can audit it rather than trust a black box.

Anatomy of a Viash component

Config + script + tests, built by Viash, into executables, Docker, Nextflow

Each component carries its own metadata, runtime, and tests. The Viash runtime turns that triple into the artefacts you need: a CLI executable, a Docker image, a Nextflow workflow module and a test report.

A

Viash component

Viash config
config.vsh.yaml
  • Functionality metadata
    • Component info
    • List of arguments
    • List of resources
  • Runner-specific metadata
    • Executable or Nextflow settings
  • Engine-specific metadata
    • Specify software dependencies
Script
>_ py R js
  • Supports many scripting languages, e.g. Bash, Python, R, and JavaScript
  • Argument parser is auto-generated
Unit test(s)
>_ py R
  • Add one or more unit tests to a component
  • Unit test can be written in a different language
  • Unit tests can be shared across components
B

Viash runtime

Build standalone executable
Build Docker image
Build Nextflow workflow
Run unit test(s)
C

Outputs

Executable
  • CLI and --help are generated by Viash
  • Runs natively or inside Docker container
Docker image

Software dependencies are translated into a Dockerfile

Nextflow workflow

Can be used as a standalone workflow or as a dependency

Test results

Detect errors early with continuous integration

Why this foundation matters

What you get because Viash is underneath

The framework is what lets us promise reproducibility, portability and audit-readiness without shipping you a black box.

Same answer, two years later

Every component is version-pinned and containerised. Re-run the same workflow next year, get the same result, with the audit trail captured at the run.

Portable to your stack

The same Viash component builds to a CLI executable, a Docker image and a Nextflow module. Run it natively, in Docker, or as part of a Nextflow pipeline. The choice is yours, not ours.

Open by default

Viash is open-source under an OSI-aligned licence and community-maintained. You can audit the framework end to end. We use it; the rest of the field can too.

Open assets

Audit it. Use it. Contribute back.

Framework, paper, source. Everything we build on is open. Here's where to find it.

Start a conversation