Documentation

The stable documentation is available on Read the Docs: nmrglue-ng.readthedocs.io/stable/.

Why nmrglue-ng?

nmrglue has provided the NMR community with a lightweight and powerful Python toolkit for reading, writing, converting, processing, and analysing NMR data for more than a decade.

Its broad support for NMR formats and its NumPy-oriented API make it particularly valuable for scientific scripting and interoperability.

nmrglue-ng builds on that work rather than replacing it, and does not seek to compete with the historical project.

One motivation for the fork was a concrete downstream need: SpectroChemPy, whose NMR support depends on nmrglue, needed a base that is actively maintained and kept compatible with current Python, NumPy, and SciPy releases. Providing that up-to-date base is part of why nmrglue-ng exists.

The main goals of this fork are:

• preserve compatibility with the established nmrglue API wherever practical;

• maintain and test existing NMR file-format support;

• fix confirmed bugs and scientific inconsistencies;

• provide reproducible test datasets for critical I/O operations;

• maintain compatibility with current Python, NumPy, and SciPy releases;

• keep packaging and continuous integration up to date;

• maintain and continuously build the documentation;

• establish a clear process for reviewing contributions as the project grows;

• document important scientific and API decisions.

Large API redesigns are explicitly not an initial objective.

Compatibility

Backward compatibility with existing nmrglue code is a primary goal.

Existing code should, whenever possible, continue to work unchanged:

import nmrglue as ng

dic, data = ng.bruker.read("experiment")

The distribution name and Python import name are intentionally distinct:

distribution: nmrglue-ng
import:       nmrglue

Compatibility cannot be guaranteed indefinitely when correcting scientifically incorrect behaviour. Such changes will be documented, tested, and introduced through an explicit compatibility policy.

Scientific scope

The project retains nmrglue’s established scope, including:

• reading and writing NMR data;

• vendor and interchange file formats;

• conversion between formats;

• spectral processing;

• unit and axis conversion;

• peak picking and fitting;

• low-memory access to large datasets.

Scientific correctness and preservation of acquisition and processing metadata are considered part of the public behaviour of the library, not implementation details.

Development priorities

The initial development effort focuses on maintenance rather than redesign.

Relationship with upstream nmrglue

nmrglue-ng is a continuation of nmrglue, not a replacement for it. It retains the complete Git history of nmrglue and acknowledges the original authors and all contributors whose work forms the basis of this project.

Where practical, fixes developed here that are applicable to the original nmrglue project may also be proposed upstream through pull requests or reported as issues in the original repository. This may include bug fixes, regression tests, documentation improvements, compatibility updates, and clarifications of scientific or format-specific behaviour.

Before opening an upstream contribution, the proposed change will be reviewed for relevance to the original project, compatibility with its maintenance policy, and consistency with its existing API and scope. Changes that are specific to nmrglue-ng, or that depend on its independent development direction, will remain in this project.

Likewise, useful upstream developments may be incorporated into nmrglue-ng with their original authorship and attribution preserved.

The intention is not to obscure or replace the history of nmrglue, but to continue building on it transparently and, where appropriate, to contribute improvements back to the original project.

Contributions

Contributions are welcome, particularly:

• reproducible bug reports;

• regression tests;

• small NMR datasets that can legally be redistributed for testing;

• fixes for file-format readers and writers;

• documentation improvements;

• scientific validation against vendor software or documented format specifications;

• performance improvements accompanied by correctness tests and benchmarks.

Changes affecting scientific conventions or public API semantics may require a short design discussion before implementation.

As the project develops, contribution and maintenance practices may evolve to make it easier for additional maintainers to participate.

Installation

The current stable release is published on PyPI: nmrglue-ng. Install it with pip:

python -m pip install nmrglue-ng

It is also available from the spectrocat Conda channel:

mamba install -c spectrocat -c conda-forge nmrglue-ng

Do not install nmrglue and nmrglue-ng in the same environment: both distributions provide the nmrglue Python package. See INSTALL.txt for source and development installation options.

The distribution is named nmrglue-ng, while the Python import remains:

import nmrglue as ng

Citation and attribution

nmrglue-ng is derived from nmrglue, created by Jonathan J. Helmus and contributors.

Users should continue to acknowledge the original nmrglue work where appropriate:

Jonathan J. Helmus and Christopher P. Jaroniec,

Nmrglue: an open source Python package for the analysis of multidimensional NMR data,

Journal of Biomolecular NMR (2013).

A citation specific to nmrglue-ng will be provided if and when an appropriate archival release or publication becomes available.

License

nmrglue-ng is distributed under the BSD 3-Clause License inherited from nmrglue.

The original copyright notices and contributor history are preserved.

See LICENSE.txt for details.

nmrglue-ng is an independent continuation of nmrglue.

Its objective is deliberately conservative: preserve a useful and established NMR toolkit while making its maintenance, testing, documentation, and evolution sustainable.