Repository navigation
What's changed
This release concludes the refineGEMs 2.0 prerelease series. The highlights below summarise the changes from 1.x. Note that this is not a complete list (see the link to the full changelog at the end).
refineGEMs 2.0 brings a major restructuring of the toolbox, redesigned gap-filling workflows, and expanded support for model curation, growth analysis, and media handling.
🆕 Highlights
- Redesigned gap-filling workflows using KEGG, BioCyc, or genome annotations with DIAMOND searches against SwissProt or a user-defined database. Reports track added genes and reactions and retain unresolved cases for manual curation.
- Energy-generating cycle detection and correction, including a greedy approach for proposing reaction changes.
- Expanded media handling through dedicated classes, updated database contents, and YAML configurations for combining and modifying media.
- Growth analysis and reporting covering simulations across models and media, auxotrophy tests, elemental-source tests, minimal media, and supplementation.
- Reorganised Python API and command-line interface, with commands grouped under analyse, curate, database, media, refine, and setup.
💻 Installation and usability
- Tested support for Python 3.10–3.12.
- Added conda support through Bioconda.
- Optional dependencies can be installed separately or together using pip install "refineGEMs[optional]".
- Docker support includes runtime extras and optional connected tools, with a build option to omit those tools.
- Updated documentation, CLI examples, installation instructions, and third-party license information.
🐛 Fixes and robustness improvements
- Improved compatibility with CarveMe models in growth tests.
- More flexible GFF parsing and improved gene-product mapping.
- Fixes to duplicate-metabolite annotation handling.
- More robust KEGG pathway group naming when pathway names are unavailable.
- Improvements to identifier mapping, gap-filling, reporting, and visualisations.
Upgrading from 1.x
Full Changelog: v1.5.0...v2.0.0