🧠 Modern Python wrappers for neuroimaging tools
- Extensive collection of neuroimaging command line tool metadata
- Generates idiomatic Python wrappers for popular neuroimaging tools
- Based on Boutiques Descriptor Schema
- Powered by Styx compiler
| Package | Status | Version | API Coverage |
|---|---|---|---|
| AFNI | Experimental | 24.2.06 |
565/611 (92.5%) |
| ANTs | Experimental | 2.5.3 |
71/113 (62.8%) |
| Connectome Workbench | Experimental | 1.5.0 |
202/202 (100% 🎉) |
| Convert3D | Experimental | 1.1.0 |
2/2 (100% 🎉) |
| FSL | Experimental | 6.0.5 |
239/310 (77.1%) |
| FreeSurfer | Experimental | 7.4.1 |
713/806 (88.5%) |
| Greedy | Experimental | 1.0.1 |
1/1 (100% 🎉) |
| MRTrix3 | Well tested | 3.0.4 |
115/121 (95.0%) |
| MRTrix3Tissue | Well tested | 5.2.8 |
1/1 (100% 🎉) |
| NiftyReg | Experimental | 1.4.0 |
7/7 (100% 🎉) |
📊 API Coverage Explained
- Represents the percentage of tool binaries with available NiWrap descriptors
- Does not indicate descriptor completeness
- 100% coverage isn't always necessary (e.g., some FSL/AFNI utilities have Python stdlib equivalents)
- To improve coverage:
- Add missing descriptor in
descriptors/- Mark irrelevant binaries as
"status": "ignore"inpackages/
niwrap/
├── descriptors/ # Boutiques descriptors
├── schemas/ # JSON schema for Boutiques descriptors
├── python/ # Generated niwrap Python package
├── extraction/ # Source metadata extraction
└── packages/ # Package-specific metadata
Install the niwrap Python package to use the generated wrappers:
pip install niwrapSee the NiWrap Python package readme for more details.
Contributions are welcome! See CONTRIBUTING.md for ways to get started.
The NiWrap project itself, including all metadata and generated Python wrappers, is licensed under the MIT License. See the LICENSE file for details.
⚠️ Important: The neuroimaging tools wrapped by NiWrap each have their own licenses. Using NiWrap does not grant any rights to use these tools beyond what their respective licenses allow. Users are responsible for complying with the licenses of the underlying tools they use through NiWrap.
For support, please open an issue in the GitHub issue tracker.