Nmrglue is an open-source Python library designed to read, write, and process multidimensional Nuclear Magnetic Resonance (NMR) data. Its core tutorial showcases how it bridges the gap between different vendor software formats, drastically simplifying macromolecular (protein, nucleic acid, and solid-state) NMR data workflows. Key Capabilities of Nmrglue
Instead of relying on rigid, closed-source legacy software, Nmrglue allows you to load raw data files directly into standard Python workflows as a tuple containing a parameter dictionary and a NumPy array.
Format Conversion: It reads, writes, and converts data seamlessly between Varian/Agilent, Bruker, NMRPipe, Sparky, SIMPSON, and the Rowland NMR Toolkit.
NumPy-Backed Processing: Because the spectrum loads as a native NumPy ndarray, you can slice, roll, and manipulate huge multi-dimensional arrays using Python speeds.
Advanced Analysis: The library provides built-in modules for linear prediction, automated peak picking, multidimensional lineshape fitting, and peak integration. Core Tutorial Workflow
The standard Nmrglue documentation tutorial highlights how to interact with macromolecular files (such as 2D and 3D data formats like .ft2 or Bruker processed directories) through Python. 1. Loading Data and Dictionaries
Loading an NMRPipe file, for instance, requires only a single line. The file is split into its metadata properties and raw spectral intensity values:
import nmrglue as ng # Load an NMRPipe 2D spectrum dic, data = ng.pipe.read(“test.ft2”) print(data.shape) # View spectral grid size/dimensions Use code with caution. 2. Mastering the Unit Conversion Object (make_uc)
One of the most practical parts of the tutorial is translating array indices (like point 512) into meaningful scientific chemical shifts (ppm or Hz).
# Create unit conversion objects for both axes of a 2D spectrum uc0 = ng.pipe.make_uc(dic, data, dim=0) uc1 = ng.pipe.make_uc(dic, data, dim=1) # Find the exact array index closest to a target chemical shift (e.g., 7.5 ppm) index_target = uc1.f(7.5, “ppm”) Use code with caution. 3. Data Slicing and Plotting
You can use matplotlib to isolate exact 1D slices from 2D or 3D macromolecular spectra—perfect for tracking small peak shifts in protein dynamics or titration studies. nmrglue – A module for working with NMR data in Python.
Leave a Reply