Evaluation of separation in gradient elution ion chromatography by combining several retention models and objective functions

In this work, three different methods for modeling of gradient retention were combined with several optimization objective functions in order to find the most appropriate combination to be applied in ion chromatography method development. The system studied was a set of seven inorganic anions (fluoride, chloride, nitrite, sulfate, bromide, nitrate, and phosphate) with a KOH eluent. The retention modeling methods tested were multilayer perceptron artificial neural network (MLP-ANN), radial-basis function artificial neural network (RBF-ANN), and retention model based on transfer of data from isocratic to gradient elution mode. It was shown that MLP retention model in combination with the objective function based on normalized retention difference product was the most adequate tool for optimization purposes.     

Interpretation of high-throughput liquid chromatography mass spectrometry data for quality control analysis and analytical method development

An approach to rapidly process and interpret high-throughput liquid chromatography mass spectrometry data is presented. This approach applies an in-house developed computer application to process LC-MS report files containing spectral and chromatographic data from four different detectors (i.e. electrospray positive ionization, electrospray negative ionization mass spectrometry, UV absorption, and evaporative light scattering detection). Properties characteristic of detection and chromatographic retention are extracted and populated into a database. Approaches to applying this analytical information database for quality control analysis of ca. 400,000 samples are presented. Compound quality assessment methods employing average purity and detection data fields are compared to methods employing multiple quality control criteria (e.g. detection, purity, retention, and signal to noise). Structural similarity searches were applied with the analytical information database to identify compounds that may be undetectable by electrospray mass spectrometry. In addition, an approach to applying the database to aid in the selection of analytical detection and chromatography conditions for rapid analytical method development is also discussed.     

MassUntangler: a novel alignment tool for label-free liquid chromatography-mass spectrometry proteomic data

Liquid chromatography-mass spectrometry (LC-MS) has become an important analytical tool for quantitative proteomics and biomarker discovery. In the label-free differential LC-MS approach computational methods are required for an accurate alignment of peaks extrapolated from the experimental raw data accounting for retention time and m/z signals intensity, which are strongly affected by sample matrix and instrumental performance. A novel procedure "MassUntangler" for pairwise alignment has been developed, relying on a pattern-based matching algorithm integrated with filtering algorithms in a multi-step approach. The procedure has been optimized employing a two-step approach. Firstly, low-complexity LC-MS data derived from the enzymatic digestion of two standard proteins have been analyzed. Then, the algorithm's performance has been evaluated by comparing the results with other achieved using state-of-the-art alignment tools. In the second step, our algorithm has been used for the alignment of high-complexity LC-MS data consisting of peptides obtained by an Escherichia coli lysate available from a public repository previously used for the comparison of other alignment tools. MassUntangler gave excellent results in terms of precision scores (from 80% to 93%) and recall scores (from 68% to 89%), showing performances similar and even better than the previous developed tools. Considering the mass spectrometry sensitivity and accuracy, this approach allows the identification and quantification of peptides present in a biological sample at femtomole level with high confidence. The procedure's capability of aligning LC-MS data previously corrected for distortion in retention time has been studied through a hybrid approach, in which MassUntangler was interfaced with the OpenMS TOPP tool MapAligner. The hybrid aligner yielded better results, showing that an integration of different bioinformatic approaches for accurate label-free LC-MS data alignment should be used.     

“PRISMA” Model for Computer-Aided HPLC Mobile Phase Optimization Based on an Automatic Peak Identification Approach

Abstract

The tripartite “PRISMA” optimization model, as part of the “PRISMA” system, includes all possible solvent combinations between 1–4 solvents, with a possible fifth one as modifier. The solvent composition is characterized by the solvent strength (S_T) and the selectivity points (P_S).

At a constant S_T the correlation between the P_S and the retention data (horizontal function) can be described by a quadratic function. For constant P_S the solvent strengths and retention data correlate (vertical function) with a logarithmic function. These correlations are used to formulate a mathematical model for the dependence of retention times (capacity factor) on the mobile phase composition. Unknown compounds are estimated in the mathematical model from a sequence of standard chromatograms after having identified individual peaks by an automatic procedure. Only retention times, relative peak areas, and information about the mobile phase compositions are required as input for the identification approach. The approach involves a combination of statistical methods which exploit both the basic properties of retention data and the mathematical relation between retention data, selectivity points, and solvent strength as derived from the “PRISMA” model. Diagnostic information for checking the identification is generated as a by-product. The mathematical model completed by the estimated constants predicts the expected retention times for each possible mobile phase combination. Peak start and peak end times are predicted in a way similar to the retention times, once the identification has been performed. The most important aspects of a chromatogram can thus be predicted for arbitrary mobile phases.

The separation quality of predicted chromatograms is assessed by the chromatographic response function (CRF). The optimal mobile phase combination is that which theoretically generates the chromatogram with the maximal CRF value. This optimal composition is found by a simple mathematical procedure, which maximises the CRF in dependence upon the mobile phase combination. The optimum found is a local one if the starting set of chromatograms contains no variation of the solvent strength, and a global one if, in the set of starting chromatograms, the solvent strength is varied in a suitable way. Recommendations for the starting position are given.

Twelve measurements are necessary for a local optimum, and 15 for the global one. To increase the accuracy, six measurements at three different solvent strength levels are proposed. Generally the highest and the lowest solvent strength level differ by ±(5)% from the middle level.

This strategy is also relevant when modifiers are used in constant amounts. The chromatographic behavior of substances to be separated can be predicted with 1% accuracy from correlations of k' values and selectivity points. Based on these relationships, an automatical mobile phase optimization strategy for isocratic separations is suggested with the “PRISMA” model.     

Column selectivity from the perspective of the solvation parameter model

The solvation parameter model is a useful tool for delineating the contribution of defined intermolecular interactions to retention of neutral molecules in separation systems based on a solute equilibrium between a gas, liquid or fluid mobile phase and a liquid or solid stationary phase. The free energy for this process is decomposed into contributions for cavity formation and the set up of intermolecular interactions identified as dispersion, electron lone pair, dipole-type and hydrogen bonding. The relative contribution of these interactions is indicated by a series of system constants determined by the difference of the defined interaction in the two phases. The interpretation of these system constants as a function of experimental factors that affect retention in the chromatographic system provides the connection between relative retention (selectivity) and the control variables for the separation system. To aid in the understanding of these processes we perform an analysis of system constants for gas chromatography, liquid chromatography, supercritical fluid chromatography and micellar electrokinetic chromatography as a function of different experimental variables as a step towards gaining a theoretical understanding of selectivity optimization for method development.     

Histidine as a dipolar eluent component in cation chromatography: II. Prediction of retention data for alkaline and alkaline-earth ions

The utility of cation chromatography has been developed by the application of -histidine as a multiprotic and dipolar (zwitterionic) eluent component. The method simplifies the cation analysis. The chromatographic characteristics of this system were studied in detail with a view to determining the selectivity and the mechanism by which the cations (Na⁺, K⁺, Mg²⁺, Ca²⁺) are retained. Complete separations were observed in the isocratic run over the eluent concentration range 3.0–6.0 mM at pH below 2.0. Sensitive detection was achieved using suppressed conductivity at the pH of isoelectric point of the histidine. Retention equations are derived for mono- and divalent cations eluted from ion-exchange separation column with multiple ionic eluents. The theory is based on the extension of ion-exchange equilibrium by protonation equilibria. The selectivity data for analyte and eluent species are determined using the model from the experimental retention data by computer-assisted iterative calculations. The model was utilized to predict retention data. The results in three-dimensional retention surfaces together with species distribution graphs are presented.     

Retention prediction and computer-assisted optimization for the separation of PTH-amino acids in isocratic reversed-phase liquid chromatography

Summary Retention prediction of phenythiohydantoin amino acid derivatives in isocratic reversed-phase liquid chromatography was investigated. The predicted retention data of all derivatives were evaluated by comparing them with actually measured retention data. Excellent agreements between these data were found. The optimized conditions to separate overlapping components can also be predicted using the developed computer-assisted optimization system with the concept of retention prediction.     

Piecewise hypersphere modeling by particle swarm optimization in QSAR studies of bioactivities of chemical compounds

As the structural diversity in a quantitative structure-activity relationship (QSAR) model increases, constructing a good model becomes increasingly difficult, and simply performing variable selection might not be sufficient to improve the model quality to make it practically usable. To combat this difficulty, an approach based on piecewise hypersphere modeling by particle swarm optimization (PHMPSO) is developed in this paper. It treats the linear models describing the sought-for subsets as hyperspheres which have different radii in the data space. According to the attribute of each hypersphere, all compounds in the training set are allocated to hyperspheres to construct submodels, and particle swarm optimization (PSO) is applied to search the optimal hyperspheres for finding satisfactory piecewise linear models. A new objective function is formulated to determine the appropriate piecewise models. The performance is assessed using three QSAR data sets. Experimental results have shown the good performance of this technique in improving the QSAR modeling.     

Predictive Optimization of the Separation of Phenylurea Pesticides using Ternary Mobile Phase Gradients in Reversed-Phase HPLC

Abstract

The method for optimization of ternary mobile phase gradients in HPLC developed earlier is illustrated on a practical example of reversed-phase separation of phenylurea pesticides. Predictive calculation methods were used to optimize subsequently “solvent strength”, “selectivity” and “combined solvent strength - selectivity” gradients. The merits of the three types of ternary gradients for this specific separation problem are discussed to elucidate the basic principles of the optimization strategy. The predicted retention data are compared with the experimental results and the limitations of the approach are shown. The present method makes it possible to reduce significantly the number of experiments and calculations necessary to perform the optimization in comparison to a systematic search of the optimized parametr space.     

Computer-assisted modelling and optimisation of reversed-phase high-temperature liquid chromatographic (RP-HTLC) separations

The use of high temperatures (above 100 °C) in reversed-phase liquid chromatography (RP-HTLC) has opened up novel and enhanced applications for this essential separation technique. Although the favourable effects of temperature on LC have been extensively studied both theoretically and practically, its potential application to method development has barely been investigated. These favourable effects include enhanced speed, efficiency, resolution and detectability, as well as changes in selectivity, especially for polar and ionisable compounds, and the emergence of new options such as temperature programming and the concomitant use of solvent and temperature gradients, green separations, and so on. The recent availability of silica-based columns that routinely support high temperatures in addition to more conventional temperature-resistant columns (based on graphitised carbon, polymers and zirconium oxide) and dedicated column ovens that allow accurate temperature control up to 200 °C makes it possible to conceive of RP-HTLC as a routine separation technique in the modern analytical laboratory. On the other hand, the addition of temperature as a new optimisable parameter to RPLC adds further complexity to method development. Thus, new computer-assisted optimisation tools that extend the capabilities of current computer-assisted tools are being specifically developed for this type of separation. A new specially developed computer-assisted method development (CAMD) tool is presented herein, and its efficiency is demonstrated. This CAMD is based on the development of a rugged retention model for peaks, allowing the simulation of any kind of RP-HTLC separation, including isocratic, linear, curved, multilinear and stepwise gradients of solvent composition concomitant with constant, linear and multilinear temperature gradients. Both the retention models and the unattended optimisation of separations are driven by evolutionary algorithms, thus providing negligible-cost, rapid, highly efficient, and user-friendly optimisation processes.     

Modelling the gas chromatographic separation of multicomponent samples on a system of three open-tubular columns coupled in series

A procedure was developed for modelling the gas chromatographic separation on a system of columns coupled in series. This procedure can be used for computer based optimization of lengths and order of serially coupled columns at isothermal conditions. The sample component retention factors and column resistances needed for the model can easily be measured on each individual column. The proposed procedure was verified by analyzing a 52 component hydrocarbon mixture on three columns of different polarity coupled in series by various column orders. Good agreement between experimental and calculated retention data was found. The procedure is also well suited for optimization of the chromatographic selectivity by coupling columns with different selectivity in series.Dedicated to Professor Dr. h.c. mult. J.F.K. Huber on the occasion of his 70th birthday     

Microcomputer-assisted liquid chromatographic separation system: application to toxic compounds identification in poisoned human fluids

The computer-assisted liquid chromatographic system (MCASYST) is developed for automated identification and analysis. The system has six main functions: retention prediction system, liquid chromatographic data base system, automated identification system, optimization of separation conditions, data loading program from UV multichannel detector, and UV spectral data base system. The performance and potential of this MCASYST system is evaluated for toxic compounds identification in poisoned human fluids.     

Combined use of algorithms for peak picking, peak tracking and retention modelling to optimize the chromatographic conditions for liquid chromatography-mass spectrometry analysis of fluocinolone acetonide and its degradation products

A strategy for rapid optimization of liquid chromatography column temperature and gradient shape is presented. The optimization as such is based on the well established retention and peak width models implemented in software like e.g. DryLab and LC simulator. The novel part of the strategy is a highly automated processing algorithm for detection and tracking of chromatographic peaks in noisy liquid chromatography-mass spectrometry (LC-MS) data. The strategy is presented and visualized by the optimization of the separation of two degradants present in ultraviolet (UV) exposed fluocinolone acetonide. It should be stressed, however, that it can be utilized for LC-MS analysis of any sample and application where several runs are conducted on the same sample. In the application presented, 30 components that were difficult or impossible to detect in the UV data could be automatically detected and tracked in the MS data by using the proposed strategy. The number of correctly tracked components was above 95%. Using the parameters from the reconstructed data sets to the model gave good agreement between predicted and observed retention times at optimal conditions. The area of the smallest tracked component was estimated to 0.08% compared to the main component, a level relevant for the characterization of impurities in the pharmaceutical industry.     

亲水模式下流动相pH值对磷酸化肽在Click OEG-CD材料上富集选择性的影响

以标准蛋白质α-酪蛋白的酶解液作为研究对象,考察流动相pH值对磷酸化肽在Click OEG-CD材料上富集选择性的影响。首先以磷酸苯二钠作为模型化合物考察流动相pH值对其在Click OEG-CD材料上的保留影响,结果表明当pH值低于磷酸根的pK_a值时,磷酸苯二钠难以电离,与材料的离子交换作用较弱,因而保留也较弱。然后在亲水模式下流动相pH值分别为2, 4, 6时考察Click OEG-CD材料对α-酪蛋白的酶解液中磷酸化肽的富集选择性影响。结果表明,当流动相pH为2时,磷酸化肽不能被材料富集;当pH为4时,磷酸化肽能够被富集,而且洗脱窗口较窄;当pH为6时,磷酸化肽也能够被富集,但是洗脱窗口较宽。因此适合亲水模式下富集磷酸化肽的流动相pH值为4。本研究结果能够为今后将Click OEG-CD材料更好的应用于磷酸化肽富集提供有意义的借鉴。     

Development of an inorganic cations retention model in ion chromatography by means of artificial neural networks with different two-phase training algorithms

This paper describes development of artificial neural network (ANN) retention model, which can be used for method development in variety of ion chromatographic applications. By using developed retention model it is possible both to improve performance characteristic of developed method and to speed up new method development by reducing unnecessary experimentation. Multilayered feed forward neural network has been used to model retention behaviour of void peak, lithium, sodium, ammonium, potassium, magnesium, calcium, strontium and barium in relation with the eluent flow rate and concentration of methasulphonic acid (MSA) in eluent. The probability of finding the global minimum and fast convergence at the same time were enhanced by applying a two-phase training procedure. The developed two-phase training procedure consists of both first and second order training. Several training algorithms were applied and compared, namely: back propagation (BP), delta-bar-delta, quick propagation, conjugate gradient, quasi Newton and Levenberg-Marquardt. It is shown that the optimized two-phase training procedure enables fast convergence and avoids problems arisen from the fact that every new weight initialization can be regarded as a new starting position and yield irreproducible neural network if only second order training is applied. Activation function, number of hidden layer neurons and number of experimental data points used for training set were optimized in order to insure good predictive ability with respect to speeding up retention modelling procedure by reducing unnecessary experimental work. The predictive ability of optimized neural networks retention model was tested by using several statistical tests. This study shows that developed artificial neural network are very accurate and fast retention modelling tool applied to model varied inherent non-linear relationship of retention behaviour with respect to mobile phase parameters.     

Retention process in reversed phase TLC systems with polar bonded stationary phases

The retention of a solute in RP chromatography is a very complex process which depends on many factors. Therefore, the study of the influence of a mobile phase modifier concentration on the retention in different reversed phase chromatographic systems is very important for understanding the rules governing retention and mechanisms of substance separation in a chromatographic process. Composition changes and the nature of mobile phases enable tuning of the separated analytes' retention over a wide range of retention parameters and optimization of the chromatographic process as well. Optimization of the chromatographic process can be achieved by several different methods; one of them is the so-called interpretative strategy. The key approach adopted in this strategy is the implementation of adequate retention models that couple the retention of solute with the composition of a mixed mobile phase. The use of chemically bonded stationary phases composed of partially non-bonded silica matrix and organic ligands bonded to its surface in everyday chromatography practice leads to questions of the correct definition of the retention model and the dominant retention mechanism in such chromatographic systems. The retention model for an accurate prediction of retention factor as a function of modifier concentration and the heterogeneity of the adsorbent surface should be taken into consideration. In this work the influence of mobile-phase composition on the retention of sixteen model substances such as phenols, quinolines, and anilines used as test analytes in different RP-TLC systems with CN-, NH2-, and Diol-silica polar bonded stationary phases has been studied. The aim of this study is to compare the performance of three valuable retention models assumed as the partition, adsorption/partition, and adsorption mechanism of retention. All the models were verified for different RP-TLC systems by three statistical criteria. The results of investigations presented in this work demonstrate that the best agreement between the experimental and calculated Rf values was obtained by the use of new-generation retention models, which assume heterogeneity of adsorbent surface. The results reported here show that heterogeneity of the adsorbent surface may be important in analysis of the elution process in liquid chromatography. Consideration of the goodness of fit for the experimental data to the examined retention models is in conformity with the adsorption mechanism of retention on all polar bonded stationary phases in most eluent systems for most investigated compounds.     

Prediction of protein retention times in gradient hydrophobic interaction chromatographic systems

A two-step methodology has been developed for the prediction of protein retention time in linear-gradient HIC systems. Isocratic retention parameters were determined from ln(k')-salt concentration plots for a number of commercially available proteins with a range of properties. Quantitative structure property relationship (QSPR) models based on a support vector machine (SVM) approach were generated for predicting isocratic retention parameters for proteins not included in the model generation. The predicted parameters were then used to calculate protein gradient retention times and the results indicate that this approach is well suited for predicting experimental gradient retention data. The approach presented in this paper may have implications for HIC methods development at both the bench and process scales.     

Retention profiles and mechanism of anion separation on latex-based pellicular ion exchanger in ion chromatography

A retention model based on stoichiometric approach has been developed in order to describe analyte retention of anions on latex-based pellicular ion exchanger. The chromatographic process entails two stepwise and complex equilibria, first is ion-pair forming of analyte or eluent ion with ion-exchange sites under the effect of electrostatic forces due to the sulfonic layer behind the aminated functional groups of stationary phase. Second component is the ion-exchange between the analyte and eluent ions. As a new parameter of the fractional electrostatic coefficient of the ion exchange capacity was introduced to develop retention profiles of anions. Analysis of the dependence of the capacity factors on the eluent concentrations at different values of fractional coefficient shed light on the possible complex mechanism. Extensive experimental retention data were obtained for 14 anions (formate, acetate, propionate, pyruvate, lactate, chloride, nitrate, oxalate, malonate, succinate, tartarate, fumarate, maleate, sulphate) using hydroxide eluents of varying concentration. The ion-pair formation and ion-exchange selectivity constants for analyte and eluent species are determined using derived retention equation from experimental data by nonlinear iterative calculation. The model was utilized to predict retention data under elution conditions of practical importance. The predicted and obtained retention factors are in good agreement, which confirms the predictive power of the model.     

乳制品中兽药多种类残留的液相色谱-质谱分析研究进展

近年来,乳制品安全问题备受关注,兽药残留作为化学危害的一个重要部分使得针对其检测技术的研究一直是乳制品安全分析的热点领域,液相色谱-质谱技术因在灵敏度和选择性方面的优势已成为目前兽药残留分析的主流技术。基于液相色谱-质谱技术的兽药残留分析趋势已逐渐在向多种类、多组分发展,但由于兽药各种类之间在理化性质、残留状态、限量要求等方面差异较大,导致在样品前处理和仪器分析环节存在一定的技术困难。为进一步了解该技术领域的研究进展,该文从样品前处理、色谱-质谱检测和基质效应等方面对近几年采用液相色谱-质谱技术测定乳制品中兽药多种类残留的国内外文献进行了综述。     

Detection and quantification of benzodiazepines and Z-drugs in human whole blood, plasma, and serum samples as part of a comprehensive multi-analyte LC-MS/MS approach

For the first time, a liquid chromatography-tandem mass spectrometry (LC-MS/MS) multi-analyte approach based on a simple liquid–liquid extraction was developed and validated for fast target screening and quantification of benzodiazepines and Z-drugs in case of driving ability and crime responsibility in the three most important biosamples whole blood, plasma, and serum. Whole blood, plasma, and serum (500 μL each) were extracted twice at pH 7.4 and at pH 10 with ether/ethyl acetate (1:1). Separation, detection, and quantification were performed using LC-MS/MS with electrospray ionization in positive mode. The method was validated with respect to selectivity, ion suppression/enhancement of co-eluting analytes, matrix effects, recovery, process efficiency, accuracy and precision, stabilities, and limits of detection and quantification. For accuracy and precision, full calibration was performed with ranges from subtherapeutic to toxic concentrations. The presented LC-MS/MS approach as part of a universal multi-analyte concept for over 100 drugs was applicable for selective detection as well as accurate and precise quantification in whole blood, plasma, and serum. The approach was selective, sensitive, accurate, and precise for 16 of the 19 tested drugs in whole blood, 18 in plasma, and 17 in serum. Only semiquantitative results could be obtained for zopiclone because of its instability in all tested biosamples.