Cross-column prediction of gas-chromatographic retention of polybrominated diphenyl ethers

Current trends in chromatographic prediction using artificial intelligence and machine learning

Artificial intelligence (AI) and machine learning (ML) gained tremendous growth and are rapidly becoming popular in various fields of prediction due to their potential abilities, accuracy, and speed. Machine learning algorithms employ historical data to analyze or predict information using patterns or trends. AI and ML were most employed in chromatographic predictions and particularly attractive options for liquid chromatography method development, as they can help achieve desired results faster, more accurately, and more efficiently. This review aims at exploring various AI and ML models employed in the determination of chromatographic characteristics. This review also aims to provide deep insight into reported artificial neural network (ANN) associated techniques which maintained better accuracy and significant possibilities for chromatographic characteristics prediction in liquid chromatography over classical linear models and also emphasizes the integration of a fuzzy system with an ANN, as this integrated study provides more efficient and accurate methods in chromatographic prediction than other linear models. This study also focuses on the retention prediction of a target molecule employing QSRR methodology combined with an ANN, highlighting a more effective technique than the QSRR alone. This approach showed the benefits of combining AI or ML algorithms with the QSRR to obtain more accurate retention predictions, emphasizing the potential of artificial intelligence and machine learning for overcoming adversities in analytical chemistry.

Transfer of gas chromatographic retention data among poly(siloxane) columns by quantitative structure-retention relationships based on molecular descriptors of both solutes and stationary phases

In the present study, computational molecular descriptors of 90 saturated esters and seven poly(siloxane) stationary phases with different polarity (SE-30, OV-7, DC-710, OV-25, XE-60, OV-225 and Silar-5CP) were combined into quantitative structure-retention relationship (QSRR) models aimed at predicting the Kováts retention indices (RIs) of the solutes. The molecular descriptors (174) of the stationary phases included in the models were computed using Dragon software from poly(siloxane) oligomers made of 20 siloxane units reflecting the nominal composition of the stationary phase, whereas 439 molecular descriptors were adopted to represent the esters. Different QSRR models were generated by means of Partial Least Squares (PLS) regression to assess the accuracy of this approach in predicting the RIs of unexplored solutes both in known and external stationary phases. After calibration of each PLS model, the descriptors were selected/discarded according to their relevance, evaluated by Covariance Selection (CovSel), and the PLS models were re-built, which resulted in a noticeable improvement of their predictive ability. Firstly, all the available data were equally divided into a training and a test set; the model built on the calibration set was used to predict the RIs of the validation observations. Successively, seven diverse PLS models were created following a "leave-one-column-out" fashion procedure, each one finalized to the estimation of the RIs of the 90 esters associated with a single stationary phase, whereas the calibration model was calculated on the remaining data. All the estimated models provided successful results on the external stationary phase, and predictive performance further increased after variable selection based on CovSel analysis. The final models provided a Root Mean Square Error in Cross Validation (RMSECV) in the range 12-20, a Root Mean Square Error in Prediction (RMSEP) in the range 11-26, and Mean Absolute Percentage Errors in Prediction (MAMEPs) in the range 0.7-1.5, revealing accurate cross-column prediction. Eventually, to test the robustness of the proposed approach, the 90 solutes were equally partitioned into a calibration and a test set and two further QSSR strategies were applied. The first PLS model was calibrated on all the seven stationary phases and the RIs of the 45 external solutes in the same seven columns were simultaneously predicted. The last QSRR approach followed a "leave-one-column-out" scheme and RI of 45 test solutes on an external stationary phase was predicted by a PLS model calibrated with the data of the 45 remaining solutes and the six left stationary phases. After selection of the significant molecular descriptors, PLS regression provided RMSECV values in the range 6-19, RMSEPs in the range 10-14, and MAPEPs in the range 0.9-2.4, revealing the suitability of the approach to deduce the RI of unknown solutes in uncharted stationary phases.

Highly efficient remediation of decabromodiphenyl ether-contaminated soil using mechanochemistry in the presence of additive and its mechanism

Mechanochemistry has been proved to be an effective method to remediation of organic-contaminated sites. However, the high ball-to-powder mass ratio (C_R) limits the large-scale application of mechanochemistry. In this study, co-milling additives were introduced to enhance the mechanochemical degradation of decabromodiphenyl ether (BDE209)-contaminated soil under the condition of low C_R. Based on additive screening experiments, sodium borohydride was selected as the ideal additive to assist the mechanochemical degradation of BDE209, and the resulting removal efficiency was approximately 100% with 2 h of ball milling at a rotational speed of 550 rpm. The main degradation intermediates and degradation pathway of BDE209 were identified using gas chromatography-tandem mass spectrometry. It was proposed that the degradation of BDE209 by sodium borohydride-assisted mechanochemistry was a concurrent process of stepwise and multistage debromination. Meanwhile, the meta-bromine atom in BDE209 was more susceptible to debromination than those at the para and ortho positions. The evolution of the concentration of Br^- was monitored by ion chromatography, which revealed that reduction and oxidation both occurred in the removal of BDE209. This paper provides a new perspective for reducing the C_R in the mechanochemical remediation of BDE209-contaminated soil.

Deep Learning Based Prediction of Gas Chromatographic Retention Indices for a Wide Variety of Polar and Mid-Polar Liquid Stationary Phases

Prediction of gas chromatographic retention indices based on compound structure is an important task for analytical chemistry. The predicted retention indices can be used as a reference in a mass spectrometry library search despite the fact that their accuracy is worse in comparison with the experimental reference ones. In the last few years, deep learning was applied for this task. The use of deep learning drastically improved the accuracy of retention index prediction for non-polar stationary phases. In this work, we demonstrate for the first time the use of deep learning for retention index prediction on polar (e.g., polyethylene glycol, DB-WAX) and mid-polar (e.g., DB-624, DB-210, DB-1701, OV-17) stationary phases. The achieved accuracy lies in the range of 16–50 in terms of the mean absolute error for several stationary phases and test data sets. We also demonstrate that our approach can be directly applied to the prediction of the second dimension retention times (GC × GC) if a large enough data set is available. The achieved accuracy is considerably better compared with the previous results obtained using linear quantitative structure-retention relationships and ACD ChromGenius software. The source code and pre-trained models are available online.

Characterisation of Gas-Chromatographic Poly(Siloxane) Stationary Phases by Theoretical Molecular Descriptors and Prediction of McReynolds Constants

Retention in gas–liquid chromatography is mainly governed by the extent of intermolecular interactions between the solute and the stationary phase. While molecular descriptors of computational origin are commonly used to encode the effect of the solute structure in quantitative structure–retention relationship (QSRR) approaches, characterisation of stationary phases is historically based on empirical scales, the McReynolds system of phase constants being one of the most popular. In this work, poly(siloxane) stationary phases, which occupy a dominant position in modern gas–liquid chromatography, were characterised by theoretical molecular descriptors. With this aim, the first five McReynolds constants of 29 columns were modelled by multilinear regression (MLR) coupled with genetic algorithm (GA) variable selection applied to the molecular descriptors provided by software Dragon. The generalisation ability of the established GA-MLR models, evaluated by both external prediction and repeated calibration/evaluation splitting, was better than that reported in analogous studies regarding nonpolymeric (molecular) stationary phases. Principal component analysis on the significant molecular descriptors allowed to classify the poly(siloxanes) according to their chemical composition and partitioning properties. Development of QSRR-based models combining molecular descriptors of both solutes and stationary phases, which will be applied to transfer retention data among different columns, is in progress.

Artificial Neural Network Prediction of Retention of Amino Acids in Reversed-Phase HPLC under Application of Linear Organic Modifier Gradients and/or pH Gradients

A multi-layer artificial neural network (ANN) was used to model the retention behavior of 16 o-phthalaldehyde derivatives of amino acids in reversed-phase liquid chromatography under application of various gradient elution modes. The retention data, taken from literature, were collected in acetonitrile⁻water eluents under application of linear organic modifier gradients ( gradients), pH gradients, or double pH/ gradients. At first, retention data collected in  gradients and pH gradients were modeled separately, while these were successively combined in one dataset and fitted simultaneously. Specific ANN-based models were generated by combining the descriptors of the gradient profiles with 16 inputs representing the amino acids and providing the retention time of these solutes as the response. Categorical "bit-string" descriptors were adopted to identify the solutes, which allowed simultaneously modeling the retention times of all 16 target amino acids. The ANN-based models tested on external gradients provided mean errors for the predicted retention times of 1.1% ( gradients), 1.4% (pH gradients), 2.5% (combined  and pH gradients), and 2.5% (double pH/ gradients). The accuracy of ANN prediction was better than that previously obtained by fitting of the same data with retention models based on the solution of the fundamental equation of gradient elution.

QSRR Modeling for Metabolite Standards Analyzed by Two Different Chromatographic Columns Using Multiple Linear Regression

Modified quantitative structure retention relationships (QSRRs) are proposed and applied to describe two retention data sets: A set of 94 metabolites studied by a hydrophilic interaction chromatography system under organic content gradient conditions and a set of tryptophan and its major metabolites analyzed by a reversed-phase chromatographic system under isocratic as well as pH and/or simultaneous pH and organic content gradient conditions. According to the proposed modification, an additional descriptor is added to a conventional QSRR expression, which is the analyte retention time, tR(R), measured under the same elution conditions, but in a second chromatographic column considered as a reference one. The 94 metabolites were studied on an Amide column using a Bare Silica column as a reference. For the second dataset, a Kinetex EVO C18 and a Gemini-NX column were used, where each of them was served as a reference column of the other. We found in all cases a significant improvement of the performance of the QSRR models when the descriptor tR(R) was considered.

PredRet: prediction of retention time by direct mapping between multiple chromatographic systems

Demands in research investigating small molecules by applying untargeted approaches have been a key motivator for the development of repositories for mass spectrometry spectra and automated tools to aid compound identification. Comparatively little attention has been afforded to using retention times (RTs) to distinguish compounds and for liquid chromatography there are currently no coordinated efforts to share and exploit RT information. We therefore present PredRet; the first tool that makes community sharing of RT information possible across laboratories and chromatographic systems (CSs). At http://predret.org , a database of RTs from different CSs is available and users can upload their own experimental RTs and download predicted RTs for compounds which they have not experimentally determined in their own experiments. For each possible pair of CSs in the database, the RTs are used to construct a projection model between the RTs in the two CSs. The number of compounds for which RTs can be predicted and the accuracy of the predictions are dependent upon the compound coverage overlap between the CSs used for construction of projection models. At the moment, it is possible to predict up to 400 RTs with a median error between 0.01 and 0.28 min depending on the CS and the median width of the prediction interval ranging from 0.08 to 1.86 min. By comparing experimental and predicted RTs, the user can thus prioritize which isomers to target for further characterization and potentially exclude some structures completely. As the database grows, the number and accuracy of predictions will increase.

Remediation of polybrominated diphenyl ethers in soil using Ni/Fe bimetallic nanoparticles: influencing factors, kinetics and mechanism

Polybrominated diphenyl ethers (PBDEs) are commonly used as additive flame retardants in all kinds of electronic products. PBDEs are now ubiquitous in the environment, with soil as a major sink, especially in e-waste recycling sites. This study investigated the degradation of decabromodiphenyl ether (BDE209) in a spiked soil using Ni/Fe bimetallic nanoparticles. The results indicated that Ni/Fe bimetallic nanoparticles are able to degrade BDE209 in soil at ambient temperature and the removal efficiency can reach 72% when an initial pH of 5.6 and at a Ni/Fe dosage of 0.03 g/g. A declining trend in degradation was noticed with decreasing Ni loading and increasing of initial BDE209 concentration. The degradation products of BDE209 were analyzed by GC-MS, which showed that the degradation of BDE209 was a process of stepwise debromination from nBr to (n-1)Br. And a possible debromination pathway was proposed. At last, the degradation process was analyzed as two-step mechanism, mass transfer and reaction. This current study shows the potential ability of Ni/Fe nanoparticles to be used for removal of PBDEs in contaminated soil.