The toolbox includes 50+ preprocessing methods. A typical NIR workflow:
% Example: Preprocessing spectrum
pp = preprocess('default', 'derivat', 2, 'width', 15);
x_pre = preprocess(x, pp);
You can chain methods: detrend, normalize, standard normal variate (SNV), and then a Savitzky–Golay derivative—all without writing complex loops.
The MATLAB PLS Toolbox is not cheap, but for serious chemometric modeling, it pays for itself in time saved alone. It bridges the gap between academic flexibility and industrial robustness. If you’re fighting with noisy spectra, overlapping peaks, or calibration maintenance, this toolbox will feel like a superpower.
Have you used the PLS Toolbox for a unique application? Let me know in the comments below. matlab pls toolbox
About the author: A chemometrician who spent years clicking through commercial software before finding the light of the PLS Toolbox. Now happier, with better models.
m = sPLS_CV(X, Y);
m = sPLS_CV(X,Y,'NumComponents',10,'LambdaGrid',logspace(-4,0,20));
Yhat = predict_sPLS(m, Xnew);
In the modern landscape of data-driven science, the ability to extract meaningful information from complex, multivariate datasets is paramount. Techniques like Principal Component Analysis (PCA) and Partial Least Squares (PLS) regression have become cornerstones of chemometrics, sensory science, process analytics, and systems biology. While the core mathematical frameworks for these methods are well-established, their effective application requires robust, flexible, and validated software. Among the most influential tools in this domain is the PLS Toolbox, a comprehensive software package that operates within the MATLAB environment. Developed and maintained by Eigenvector Research, Incorporated, the PLS Toolbox has evolved over three decades from a niche academic tool into an industry-standard platform. This essay provides a long-form exploration of the PLS Toolbox, examining its historical context, core functionalities, distinctive methodological philosophy, practical applications, and its standing relative to other chemometric software.
Add sparse PLS (L1-penalized loadings) with automatic selection of: The toolbox includes 50+ preprocessing methods
The versatility of the PLS Toolbox has led to its adoption across a wide range of industries and academic fields.
Pharmaceuticals (Process Analytical Technology - PAT): In drug manufacturing, the FDA encourages real-time quality monitoring. The PLS Toolbox is used to build multivariate calibration models that predict API concentration or blend homogeneity from NIR spectra acquired directly from a mixing vessel. Its robust outlier detection is crucial for flagging abnormal process events.
Food and Agriculture: For determining fat, protein, or moisture content in meat, grain, or dairy products. The toolbox’s ability to handle MSC and derivatives corrects for physical scatter effects due to particle size or sample packing. You can chain methods: detrend, normalize, standard normal
Petrochemicals: Modeling octane number, viscosity, or distillation curves from NIR or MIR spectra of crude oil and fuels. The multiway methods are used for analyzing batch reactors.
Environmental Chemistry: PARAFAC decomposition of fluorescence EEMs to identify and quantify dissolved organic matter in water samples—a classic application that is almost impossible without dedicated software like the PLS Toolbox.
Sensory Science and Consumer Products: Relating instrumental measurements (e.g., rheology or spectroscopy) to human sensory panel scores using PLS2, which can handle multiple response variables simultaneously (e.g., sweetness, bitterness, texture).