Skip to content
RepoReady
Sign up
Menu
Docs menu

Software reuse: code others can build on

Reuse asks whether a named algorithm or package can be used and adapted beyond your paper. RepoReady does not assess it yet, and it never changes the paper’s level.

Last updated

Reuse belongs to a repository’s own software, next to the paper’s level.
Level relationshipSeparate axis · not assessed yet

A method that others can install and run on their own data travels further than the paper. The JOSS review criteria make a good checklist: documentation, installation, example usage, tests and a license.

Installation

What counts
The software installs with one standard command from a package index or the repository, with its dependencies declared.
Common gaps
Code that only runs from inside the analysis folder, with paths to your own data.
How to fix
Package the method separately from the analysis, with a pyproject.toml or an R DESCRIPTION:
pyproject.toml
[project]
name = "cellfilter"
version = "1.2.0"
description = "Quality filtering for single-cell count matrices"
requires-python = ">=3.10"
dependencies = ["numpy>=1.26", "anndata>=0.10"]
license = "MIT"

[project.optional-dependencies]
test = ["pytest"]

Documented interface and examples

What counts
A README or documentation site explains inputs, outputs and options, with a short example that runs on test data.
Common gaps
Functions described only in the Methods section of the paper.
How to fix
Document the public functions and ship a small example dataset with a worked example.

Automated tests

What counts
Tests check the core functionality on small, shipped data, ideally against known results.
Common gaps
Tests that need your full dataset or your cluster, so nobody else can run them.
How to fix
Add tests that run in seconds on shipped data, and run them on every change.

Releases and license

What counts
Numbered releases with a changelog, each archived with a DOI, under an open license. The license is checked once, under Journal policy; this area shows that result.
Common gaps
One snapshot made for the paper, with no word on what may change.
How to fix
Release with version numbers, note changes per release and archive each one, for example with Zenodo.

See where your project stands

Run the analysis on your paper or repository. Every finding names its area and links back here.

Get started