Theoretical modeling and machine learning-based data processing workflows in comprehensive two-dimensional gas chromatography-A review.

In recent years, comprehensive two-dimensional gas chromatography (GC × GC) has been gradually gaining prominence as a preferred method for the analysis of complex samples due to its higher peak capacity and resolution power compared to conventional gas chromatography (GC). Nonetheless, to fully benefit from the capabilities of GC × GC, a holistic approach to method development and data processing is essential for a successful and informative analysis. Method development enables the fine-tuning of the chromatographic separation, resulting in high-quality data. While generating such data is pivotal, it does not necessarily guarantee that meaningful information will be extracted from it. To this end, the first part of this manuscript reviews the importance of theoretical modeling in achieving good optimization of the separation conditions, ultimately improving the quality of the chromatographic separation. Multiple theoretical modeling approaches are discussed, with a special focus on thermodynamic-based modeling. The second part of this review highlights the importance of establishing robust data processing workflows, with a special emphasis on the use of advanced data processing tools such as, Machine Learning (ML) algorithms. Three widely used ML algorithms are discussed: Random Forest (RF), Support Vector Machine (SVM), and Partial Least Square-Discriminate Analysis (PLS-DA), highlighting their role in discovery-based analysis.

Tile-Based Random Forest Analysis for Analyte Discovery in Balanced and Unbalanced GC × GC-TOFMS Data Sets.

In this study, we introduce a new nontargeted tile-based supervised analysis method that combines the four-grid tiling scheme previously established for the Fisher ratio (F-ratio) analysis (FRA) with the estimation of tile hit importance using the machine learning (ML) algorithm Random Forest (RF). This approach is termed tile-based RF analysis. As opposed to the standard tile-based F-ratio analysis, the RF approach can be extended to the analysis of unbalanced data sets, i.e., different numbers of samples per class. Tile-based RF computes out-of-bag (oob) tile hit importance estimates for every summed chromatographic signal within each tile on a per-mass channel basis (m/z). These estimates are then used to rank tile hits in a descending order of importance. In the present investigation, the RF approach was applied for a two-class comparison of stool samples collected from omnivore (O) subjects and stored using two different storage conditions: liquid (Liq) and lyophilized (Lyo). Two final hit lists were generated using balanced (8 vs Eight comparison) and unbalanced (8 vs Nine comparison) data sets and compared to the hit list generated by the standard F-ratio analysis. Similar class-distinguishing analytes (p < 0.01) were discovered by both methods. However, while the FRA discovered a more comprehensive hit list (65 hits), the RF approach strictly discovered hits (31 hits for the balanced data set comparison and 29 hits for the unbalanced data set comparison) with concentration ratios, [OLiq]/[OLyo], greater than 2 (or less than 0.5). This difference is attributed to the more stringent feature selection process used by the RF algorithm. Moreover, our findings suggest that the RF approach is a promising method for identifying class-distinguishing analytes in settings characterized by both high between-class variance and high within-class variance, making it an advantageous method in the study of complex biological matrices.

Top-Down Approach to Retention Time Prediction in Comprehensive Two-Dimensional Gas Chromatography-Mass Spectrometry.

In this contribution, we describe a novel modeling approach to predicting retention times (tr) in comprehensive two-dimensional gas chromatography coupled to time-of-flight mass spectrometry (GC × GC-ToF-MS) with a particular emphasis on the second-dimension (2D) retention time predictions (2tr). This approach is referred to as a "top-down" approach in that it breaks down the complete GC × GC separation into two independent one-dimensional gas chromatography separations (1D-GC). In this regard, both dimensions, that is, first dimension (1D) and second dimension (2D) are treated separately, and the cryogenic modulator is simply considered as a second consecutive injection device. Separate 1D-GC tr predictions are performed on both dimensions using the same flow rate as the one deployed in the conventional GC × GC system. The separate tr predictions are then combined to account for the two-dimensional separation. This model was applied to 24 analytes from 2 standard mixtures (Grob Test Mix and Fragrance Materials Test Mix) and assessed across 9 GC × GC chromatographic conditions. The experimental and predicted chromatographic retention space occupations were assessed by using the convex hull approach defined by the Delaunay triangulation. The predicted percentage of space occupation corresponded favorably with the experimental values. Furthermore, the top-down approach enabled an accurate prediction of the 2tr of all investigated analytes, providing an average 2tr modeling error of 0.26 ± 0.01 s.

Distinguishing between Decaffeinated and Regular Coffee by HS-SPME-GC×GC-TOFMS, Chemometrics, and Machine Learning.

Coffee, one of the most popular beverages in the world, attracts consumers by its rich aroma and the stimulating effect of caffeine. Increasing consumers prefer decaffeinated coffee to regular coffee due to health concerns. There are some main decaffeination methods commonly used by commercial coffee producers for decades. However, a certain amount of the aroma precursors can be removed together with caffeine, which could cause a thin taste of decaffeinated coffee. To understand the difference between regular and decaffeinated coffee from the volatile composition point of view, headspace solid-phase microextraction two-dimensional gas chromatography time-of-flight mass spectrometry (HS-SPME-GC×GC-TOFMS) was employed to examine the headspace volatiles of eight pairs of regular and decaffeinated coffees in this study. Using the key aroma-related volatiles, decaffeinated coffee was significantly separated from regular coffee by principal component analysis (PCA). Using feature-selection tools (univariate analysis: t-test and multivariate analysis: partial least squares-discriminant analysis (PLS-DA)), a group of pyrazines was observed to be significantly different between regular coffee and decaffeinated coffee. Pyrazines were more enriched in the regular coffee, which was due to the reduction of sucrose during the decaffeination process. The reduction of pyrazines led to a lack of nutty, roasted, chocolate, earthy, and musty aroma in the decaffeinated coffee. For the non-targeted analysis, the random forest (RF) classification algorithm was used to select the most important features that could enable a distinct classification between the two coffee types. In total, 20 discriminatory features were identified. The results suggested that pyrazine-derived compounds were a strong marker for the regular coffee group whereas furan-derived compounds were a strong marker for the decaffeinated coffee samples.

Modeling approaches for temperature-programmed gas chromatographic retention times under vacuum outlet conditions.

This contribution evaluates the performance of two predictive approaches in calculating temperature-programmed gas chromatographic retention times under vacuum outlet conditions. In the first approach, the predictions are performed according to a thermodynamic-based model, while in the second approach the predictions are conducted by using the temperature-programmed retention time equation. These modeling approaches were evaluated on 47 test compounds belonging to different chemical classes, under different experimental conditions, namely, two modes of gas flow regulation (i.e., constant inlet pressure and constant flow rate), and different temperature programs (i.e., 7 °C/min, 5 °C/min, and 3 °C/min). Both modeling approaches gave satisfactory results and were able to accurately predict the elution profiles of the studied test compounds. The thermodynamic-based model provided more satisfying results under constant flow rate mode, with average modeling errors of 0.43%, 0.33%, and 0.15% across all the studied temperature programs. Nevertheless, under constant inlet pressure mode, lower modeling errors were achieved when using the temperature-programmed retention time equation, with average modeling errors of 0.18%, 0.18%, and 0.31% across the used temperature programs.