Nextflow (https://nextflow.io) enables scalable and reproducible scientific workflows using software containers. Nextflow uses a simple, fluent Domain Specific Language (DSL) and is backed by a thriving Open-Source community. With many community-built and industry proven sample pipelines, it is a strong option for any data science workload.
For a real-world scenario, head over to Joe Karasha’s blog series: COVID Variant Analysis on Azure using Nextflow.
Nextflow system requirements:
- Any POSIX compatible system (Linux, OS X)
- Bash 3.2 (or later)
- Java 8 (or later)
These don’t look too scary. However, if you are in a Windows environment and unable to use Windows Subsystem for Linux (WSL) due to hardware or corporate policy constraints, you’ll have to find a suitable virtual environment.
GitHub Codespaces offer the perfect solution. These virtual development environments can be spun up in several minutes or less. They are configurable and customizable, allowing us to create an image with the same set of development dependencies and tooling every time. After 30 minutes of inactivity, they are automatically spun down to save costs.
Try this out today! Head over to my Hello World! Nextflow sample template repo on GitHub. There, you can easily create your own sample repository using the template and have a Nextflow Codespace running in just a few minutes.