pylacs 1.0.1

Robust LACS offsets, uncertainties, outliers, and plots from NMR-STAR files

PyLACS : Python version Linear Analysis of Chemical Shifts(LACS)

PyLACS is a Python package for detecting and correcting chemical shift referencing errors in protein NMR spectroscopy datasets, using robust linear regression methods.
It provides both a Python API and a command-line interface (CLI) for use in research pipelines, HPC batch jobs, and interactive exploration.

---

✨ Features

• Detect and correct chemical shift referencing errors for Cα, Cβ, C, N, and H nuclei
• Robust regression methods:
  • Tukey biweight
  • Theil–Sen estimator
  • RANSAC
  • Quantile regression
  • Bayesian regression (via PyMC + ArviZ)
• Option to chose different Random Coil chemical shifts
• Outlier detection and offset estimation
• Optional plot generation (scatter, regression fits)
• Works directly with BMRB NMR-STAR (.str) files
• Provides both a CLI (pylacs) and a Python API

---

📦 Installation

Stable release from PyPI:

pip install pylacs

---

🚀 Quickstart

pylacs myfile.str --data-id my_data_id --method tukey --out results/

This will compute offsets, generate validation reports, and (if plotting is enabled) save plots in results folder

For help

pylacs --help

📖 Citation

If you use PyLACS in your research, please cite:

Wang, L., Eghbalnia, H. R., & Markley, J. L. (2005). Probabilistic approach to determining protein backbone torsion angles from NMR chemical shifts. Journal of Biomolecular NMR, 32(1), 13–22.

Baskaran, K., et al. (2025). PyLACS: Python-based Linear Analysis of Chemical Shifts pipeline. (manuscript in preparation)