Why epi4you exists¶
Oxford Nanopore’s EPI2ME Desktop gives users a convenient local environment for running and browsing nanopore workflows. That convenience is valuable, but it also means a completed analysis is spread across several concerns:
the workflow output itself,
the metadata that lets the GUI present the run coherently,
the workflow identity and versioning information, and
sometimes related software assets such as containers or workflow trees.
For everyday use on a single workstation, that is fine.
For training and deployment work, it becomes awkward.
The practical problem¶
If you need to prepare twenty laptops for a workshop, or preload a conference demo machine, copying one output directory is often not enough.
Typical problems include:
the target machine has no matching EPI2ME database entry,
the original workflow version is not obvious,
the GUI expects helper files that a plain CLI run never created,
internet access is limited or unavailable, and
rebuilding analyses from raw data would take too long.
epi4you’s answer¶
epi4you exists to make that liftover explicit and repeatable.
The software packages a run into a manifest-driven archive so the transfer is not just “copy these files and hope”. Instead the archive includes:
a typed payload description,
a file inventory,
provenance,
a lightweight integrity check, and
enough reconstructed metadata to make the imported result behave like an EPI2ME-style analysis.
Why the project is broader than the current CLI¶
The repository contains logic for several related asset classes:
Desktop analysis records,
workflow installations,
Docker/container artefacts, and
historical Nextflow CLI runs.
The currently active entry points focus on packaging a CLI Nextflow run and
importing a .2me archive, but the wider codebase reflects the broader
problem the project was created to solve: moving complete bioinformatics assets
between machines, not just moving raw result files.
Audience¶
The tool is especially well suited to:
trainers and course organizers,
field teams carrying prebuilt examples,
demo and conference operators,
support engineers reproducing representative analyses, and
labs that need to share curated EPI2ME-ready examples internally.