v1.0.0

First stable release of maldiassist (Python port of MALDIassist).

• PyPI: https://pypi.org/project/maldiassist/
• nanobind C++ acceleration for KDE hot paths with pure-Python fallback
• Win / macOS (universal2) / Linux wheels (CPython 3.9–3.13)

Full Changelog: v0.2.2...v1.0.0

v0.2.2

Full Changelog: v0.2.1...v0.2.2

v0.2.1

Full Changelog: v0.2.0...v0.2.1

v0.2.0

Full Changelog: v0.1.0...v0.2.0

v0.1.0

maldiassist v0.1.0

First release of maldiassist, a pure-Python port of the
MALDIassist R package (v0.1.3, R + Rcpp/C++)
that reproduces the same algorithms and numerical results.

Highlights

• Pure-Python implementation (numpy, scipy, pandas) — no compiler required.
• Full five-step MALDI-TOF workflow, matching the original R package:
  • Bruker Flex raw-data loading (load_maldi_spectra)
  • Preprocessing: Savitzky-Golay smoothing + SNIP/TopHat baseline (preprocess_maldi_spectra)
  • Kernel-regression peak detection incl. shoulder peaks (find_peaks / find_peaks_spectra)
  • Peak metrics and filtering (filter_peaks / filter_peaks_spectra)
  • Cohort analysis: alignment, frequent m/z, matched matrix, significance testing
  • Visualization (visualize_spectrum/spectra, heatmap_matched_matrix)

Numerical fidelity

Validated step by step against ground truth produced by running the original R
package on the TestData set, achieving an exact match within floating-point
tolerance. Kernel summation uses the same sequential accumulation order as R's
Rcpp loop to preserve tie-breaking in extremum selection.

Installation

pip install "git+https://github.com/hiows/MALDIassist-py.git"
pip install "maldiassist[viz] @ git+https://github.com/hiows/MALDIassist-py.git"  # with matplotlib

Requires Python 3.9+.