Comparison of variable selection methods in partial least squares regression

Mehmood, Tahir; Sæbø, Solve; Liland, Kristian Hovde

doi:10.1002/cem.3226

Cited by 131 publications

(77 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mean-squared error was determined using a 5-fold cross validation, and was used to select the number of vector components retained in each PLSR model. 26 Pearson's linear correlation coefficient was used to compare drift-subtracted CVs to those obtained using the conventional background subtraction approach. All statistical analyses and graphical depiction of data were carried out using GraphPad Prism 6 (GraphPad Software, Inc., La Jolla, CA) or MATLAB R2018a.…”

Section: Data Processing and Analysismentioning

confidence: 99%

“…By contrast, PLSR is a supervised dimensionality reduction method that projects both predictor and response variables to a new vector space to determine the PCs that maximize the covariance of projected structures. 26,42 As such, PLSR generally describes training data more efficiently with fewer PCs (than PCR), and output prediction is often more robust. [43][44] The DW-PLSR model is trained using data collected with the sWF and the lWF as the predictor and response, respectively.…”

Section: The Double-waveform Partial-least-squares Regression Modelmentioning

confidence: 99%

See 1 more Smart Citation

Drift Subtraction for Fast-Scan Cyclic Voltammetry Using Double-Waveform Partial-Least-Squares Regression

2019

View full text Add to dashboard Cite

Background-subtracted fast-scan cyclic voltammetry (FSCV) provides a method for detecting molecular fluctuations with high spatiotemporal resolution in the brain of awake and behaving animals. The rapid scan rates generate large background currents that are subtracted to reveal changes in analyte concentration. Although these background currents are relatively stable, small changes do occur over time. These changes, referred to as electrochemical drift, result in background-subtraction artifacts that constrain the utility of FSCV, particularly when quantifying chemical changes that gradually occur over long measurement times (minutes). The voltammetric features of electrochemical drift are varied and can span the entire potential window, potentially obscuring the signal from any targeted analyte. We present a straightforward method for extending the duration of a single FSCV recording window. First, we have implemented voltammetric waveforms in pairs that consist of a smaller triangular sweep followed by a conventional voltammetric scan. The initial, abbreviated waveform is used to capture drift information that can serve as a predictor for the contribution of electrochemical drift to the subsequent full voltammetric scan using partial-least-squares regression (PLSR). This double-waveform partialleast-squares regression (DW-PLSR) paradigm permits reliable subtraction of the drift component to the voltammetric data. Here, DW-PLSR is used to improve quantification of adenosine, dopamine, and hydrogen peroxide fluctuations occurring >10 min from the initial background position, both in vitro and in vivo. The results demonstrate that DW-PLSR is a powerful tool for evaluating and interpreting both rapid (seconds) and gradual (minutes) chemical changes captured in FSCV recordings over extended durations.

show abstract

Section: Data Processing and Analysismentioning

confidence: 99%

Section: The Double-waveform Partial-least-squares Regression Modelmentioning

confidence: 99%

Drift Subtraction for Fast-Scan Cyclic Voltammetry Using Double-Waveform Partial-Least-Squares Regression

2019

View full text Add to dashboard Cite

show abstract

“…Numerical methods, such as sRatio and VIP select variables by the amount of a specific value [23]. With sRatio, the ratio of explained variance to residual variance was calculated for each variable.…”

Section: Multivariate Data Analysismentioning

confidence: 99%

Indirect Nuclear Magnetic Resonance (NMR) Spectroscopic Determination of Acrylamide in Coffee Using Partial Least Squares (PLS) Regression

et al. 2021

View full text Add to dashboard Cite

Acrylamide is probably carcinogenic to humans (International Agency for Research on Cancer, group 2A) with major occurrence in heated, mainly carbohydrate-rich foods. For roasted coffee, a European Union benchmark level of 400 µg/kg acrylamide is of importance. Regularly, the acrylamide contents are controlled using liquid chromatography combined with tandem mass spectrometry (LC–MS/MS). This reference method is reliable and precise but laborious because of the necessary sample clean-up procedure and instrument requirements. This research investigates the possibility of predicting the acrylamide content from proton nuclear magnetic resonance (NMR) spectra that are already recorded for other purposes of coffee control. In the NMR spectrum acrylamide is not directly quantifiable, so that the aim was to establish a correlation between the reference value and the corresponding NMR spectrum by means of a partial least squares (PLS) regression. Therefore, 40 commercially available coffee samples with already available LC–MS/MS data and NMR spectra were used as calibration data. To test the accuracy and robustness of the model and its limitations, 50 coffee samples with extreme roasting degrees and blends were additionally prepared as the test set. The PLS model shows an applicability for the varieties Coffea arabica and C. canephora, which were medium to very dark roasted using drum or infrared roasters. The root mean square error of prediction (RMSEP) is 79 µg/kg acrylamide (n = 32). The current PLS model is judged as suitable to predict the acrylamide values of commercially available coffee samples.

show abstract

“…Numerical methods, such as sRatio and VIP select variables by the amount of a specific value [23]. With sRatio, the ratio of explained variance to residual variance is calculated for each variable.…”

Section: Multivariate Data Analysismentioning

confidence: 99%

Indirect Nuclear Magnetic Resonance (NMR) Spectroscopic Determination of Acrylamide in Coffee Using Partial Least Squares (PLS) Regression

Rief¹,

Felske²,

Scharinger³

et al. 2021

Preprint

View full text Add to dashboard Cite

Acrylamide is probably carcinogenic to humans (International Agency for Research on Cancer, group 2A) with major occurrence in heated, mainly carbohydrate-rich foods. For roasted coffee, a European Union benchmark level of 400 µg/kg acrylamide is of importance. Regularly, the acrylamide contents are controlled using liquid chromatography combined with tandem mass spectrometry (LC-MS/MS). This reference method is reliable and precise but laborious because of the necessary sample clean-up procedure and instrument requirements. This research investigates the possibility of predicting the acrylamide content from proton nuclear magnetic resonance (NMR) spectra that are already recorded for other purposes of coffee control. In the NMR spectrum acrylamide is not directly quantifiable, so that the aim was to establish a correlation between the reference value and the corresponding NMR spectrum by means of a partial least squares (PLS) regression. Therefore, 40 commercially available coffee samples with already available LC-MS/MS data and NMR spectra were used as calibration data. To test the accuracy and robustness of the model and its limitations, 50 coffee samples with extreme roasting degrees and blends were additionally prepared as test set. The PLS model shows an applicability for the varieties C. arabica and C. canephora, which were medium to very dark roasted using drum or infrared roasters. The root mean square error of prediction (RMSEP) is 79 µg/kg acrylamide (n=32). The PLS model is judged as suitable to predict the acrylamide values of commercially available coffee samples. On the other hand, very light roasts containing more than 1000 µg/kg acrylamide are currently not suitable for PLS prediction.

show abstract

Comparison of variable selection methods in partial least squares regression

Cited by 131 publications

References 45 publications

Drift Subtraction for Fast-Scan Cyclic Voltammetry Using Double-Waveform Partial-Least-Squares Regression

Drift Subtraction for Fast-Scan Cyclic Voltammetry Using Double-Waveform Partial-Least-Squares Regression

Indirect Nuclear Magnetic Resonance (NMR) Spectroscopic Determination of Acrylamide in Coffee Using Partial Least Squares (PLS) Regression

Indirect Nuclear Magnetic Resonance (NMR) Spectroscopic Determination of Acrylamide in Coffee Using Partial Least Squares (PLS) Regression

Contact Info

Product

Resources

About