Reproducibility

How Jia-Ming Chang’s lab at NCCU releases code and data: where to find it, what licence it carries, and the standard each new release is held to.

A result you cannot rerun is a claim, not a finding. This page states where our code and data live, what you are allowed to do with them, and the standard we hold each new release to.

Where to find it

  • Publications — each selected paper carries Code, Data and Web server buttons where those exist.
  • Research — a Data & Code list under each research direction.
  • Packages — every tool and server in one place.
  • github.com/changlabtw — the lab’s repositories.

Sequencing data underlying our papers is deposited in the public archive appropriate to the data type; accession numbers are given in the paper itself, which stays the authoritative record.

Licensing

Each repository states its own licence — check the LICENSE file before reusing code. Text and figures on this site are CC BY 4.0.

Code with no licence file is, legally, all rights reserved: nobody may reuse it, whatever the intent. A missing licence is a bug, and we treat it as one.

What a release contains

The standard for anything we publish alongside a paper:

  1. A README that names versions. Dependencies with the versions actually used, not “recent”. A reader six years from now has no way to guess.
  2. A worked example on small data. Something that runs in minutes and produces output the reader can compare against. Without it, a failed install is indistinguishable from a broken tool.
  3. A LICENSE file. See above.
  4. A container where the dependency chain is awkward. Bioinformatics pipelines rot from underneath — a Docker image is often the only thing that still runs.
  5. A tag or archived DOI at publication time. main moves; the paper’s results came from one specific state of the code.

wpSBOOT is the reference: pinned dependency versions, a Docker image, continuous integration, and a runnable example.

Something missing or broken?

Links rot because other people shut services down, not because we changed anything — this site’s external links are checked monthly, but archives and servers outside our control still disappear. If a link is dead, a dataset is not where the paper says it is, or an example does not run, please get in touch or open an issue on the repository. We would rather hear it from you than not know.