Extracting Information and Enhancing the Quality of Separation Data: A Review on Chemometrics-Assisted Analysis of Volatile, Soluble and Colloidal Samples

Abstract

Instrument automation, technological advancements and improved computational power made separation science an extremely data-rich approach, requiring the use of statistical and data analysis tools that are able to optimize processes and combine multiple outputs. The use of chemometrics is growing, greatly improving the ability to extract meaningful information. Separation–multidetection generates multidimensional data, whose elaboration should not be left to the discretion of the operator. However, some applications or techniques still suffer from the lack of method optimization through DoE and downstream multivariate analysis, limiting their potential. This review aims at summarizing how chemometrics can assist analytical chemists in terms of data elaboration and method design, focusing on what can be achieved by applying chemometric approaches to separation science. Recent applications of chemometrics in separation analyses, in particular in gas, liquid and size-exclusion chromatography, together with field flow fractionation, will be detailed to visualize the state of the art of separation chemometrics, encompassing volatile, soluble and solid (colloidal) analytes. The samples considered will range from food chemistry and environmental chemistry to bio/pharmaceutical science.

SVR Chemometrics to Quantify β-Lactoglobulin and α-Lactalbumin in Milk Using MIR

“…Two different sample partitioning techniques were employed: the Kennard–Stone algorithm (KS), and random splitting using the scikit-learn library (RS) [ 36 ]. To ensure robust model development and evaluation on the limited dataset, leave-one-out cross-validation (LOOCV) was further implemented on the full dataset (calibration and validation sets) [ 36 , 37 ].…”

“…Chemometrics utilize mathematical or statistical methods to select optimal measurement procedures to extract relevant chemical information from chromatographic and spectroscopic data. Chemometrics has emerged as a valuable tool for the interpretation and analysis of complex datasets from gas chromatography, liquid chromatography, and infrared spectroscopy [ 37 , 38 ]. Advances in technology have shown IR-based chemometrics can be used for the rapid and accurate assessment of components in food products including milk, meat, and potato.…”

Quantification of Recycled PET in Commercial Bottles by IR Spectroscopy and Chemometrics

“…The corrected NIR and MIR-ATR spectra were used to compute a PLS regression model. PLS is a consolidated chemometric method [26,32] that performs a regression using the experimental variables (in this case, IR spectra) as predictors and one or more continuous variables characterizing the samples (in this case, the label-declared percentage of r-PET) as dependent ones. The computation is carried out by calculating factors that are linear combinations of the predictors and retaining part of the information derived also from the dependent variable(s).…”

Field-Flow Fractionation in Molecular Biology and Biotechnology

“…This approach also represents a proof of concept for the further investigation of selective interactions with small molecules and probes of clinical/pharmaceutical interest [98]. Recently, a completely different approach to the FlFFF of complex samples has been proposed, based on the application of principal component analysis (PCA) and other chemometric tools [99], on the fractograms of different samples of a certain food matrix [100,101]. A recent study highlighted the role of proteins in discriminating the regional provenience of tomatoes [102].…”

“…For example, in 2015, Johann et al developed ElAsFlFFF, a separative platform that combines ElFFF and AF4 [20]. This system allows separation based on electrophoretic mobility, overcoming some of the restrictions of ElFFF such as the electrode interference, the requirement of low ionic strength Recently, a completely different approach to the FlFFF of complex samples has been proposed, based on the application of principal component analysis (PCA) and other chemometric tools [99], on the fractograms of different samples of a certain food matrix [100,101]. A recent study highlighted the role of proteins in discriminating the regional provenience of tomatoes [102].…”