Toxicological classification of urine samples using pattern recognition techniques and capillary electrophoresis

Abstract

In toxicology, hazardous substances detected in organisms may often lead to different pathological conditions depending on the type of exposure and level of dosage; hence, further analysis on this can suggest the best cure. Urine profiling may serve the purpose because samples typically contain hundreds of compounds representing an effective metabolic fingerprint. This paper proposes a pattern recognition procedure for determining the type of cadmium dosage, acute or chronic, administrated to laboratory rats, where urinary profiles are detected using capillary electrophoresis. The procedure is based on the composition of a sample data matrix consisting of areas of common peaks, with appropriate pre-processing aimed at reducing the lack of reproducibility and enhancing the potential contribution of low-level metabolites in discrimination. The matrix is then used for pattern recognition including principal components analysis, cluster analysis, discriminant analysis and support vector machines. Attention is particularly focussed on the last of these techniques, because of its novelty and some attractive features such as its suitability to work with datasets that are small and/or have low samples/variable ratios. The type of cadmium administration is detected as a relevant feature that contributes to the structure of the sample matrix, and samples are classified according to the class membership, with discriminant analysis and support vector machines performing complementarily on a training and on a test set.

Rapid detection and identification of N-acetyl-L-cysteine thioethers using constant neutral loss and theoretical multiple reaction monitoring combined with enhanced product-ion scans on a linear ion trap mass spectrometer

“…Therefore, the aim was to develop a screening method to detect unknown compounds that carry identical substructures. This approach permits the preselection of an analyte group [9]. The preselection criterion may be a specific product ion such as m/z 113 in glucuronides, [10] a specific neutral loss such as 162 Da involving a glucose substructure as present in glycosides [9], or the neutral loss of 129 Da of glutathione metabolites [11].…”

Data processing in metabolic fingerprinting by CE‐UV: Application to urine samples from autistic children

“…Unsupervised pattern recognition methods such as cluster analysis and principal component analysis (PCA) have been used to examine the structure of these datasets [13,17,21,22]. In combination with PCA for dimensionality reduction, discriminant analysis (DA) methods can be used to classify samples according to their characteristic profiles [13,17,22,23].…”

“…MEKC, applicable to both ionic and neutral species, has proved to be a useful tool in the study of different target metabolites in urine [15][16][17][18]. However, its application as a nontargeted approach has not yet been fully exploited.…”

“…In combination with PCA for dimensionality reduction, discriminant analysis (DA) methods can be used to classify samples according to their characteristic profiles [13,17,22,23]. Other methods used for the classification of CE data include neural networks [21] and support vector machines (SVMs) [17]. However, before such methods can be used for direct comparison of datasets, preprocessing is required to remove the influence of different instrumental factors and to normalise the data so that effects due to variation in concentration do not dominate the classification.…”

“…CE has recently been incorporated to the family of metabolic fingerprinting techniques used for clinical studies [11][12][13][14]. CE is an efficient and sensitive high-resolution separation technique that allows fast analysis of ionic species and the use of small samples.MEKC, applicable to both ionic and neutral species, has proved to be a useful tool in the study of different target metabolites in urine [15][16][17][18]. However, its application as a nontargeted approach has not yet been fully exploited.…”