Metabolic profiling of human blood serum from treated patients with bipolar disorder employing 1H NMR spectroscopy and chemometrics.

Metabolic profiling employing hydrogen nuclear magnetic resonance (1H NMR) spectroscopy and chemometric analysis of human blood serum samples taken from the control group (n = 25) and patients with bipolar disorder (n = 25) was performed to identify molecular changes related to the disorder and to different drug treatments: lithium (n = 15) versus other medications (n = 10). This strategy showed significant potential for exploring pathophysiological and toxicological features involved in bipolar disorder. The investigated groups (control and patients with bipolar disorder under different treatments) could be distinguished according to their metabolic profiles, and the main differential metabolites found were lipids, lipid-metabolism-related molecules (acetate, choline, and myo-inositol), and some key amino acids (glutamate, glutamine). Our results suggest that some of the 24 identified metabolites may be linked to lithium- and other-medication-provoked metabolic changes or may even be directly related to the disorder. Thus, these findings may contribute to paving the way for future studies aiming at identifying potential biomarkers for bipolar disorder.

Least-squares support vector machines and near infrared spectroscopy for quantification of common adulterants in powdered milk.

This paper proposes the use of the least-squares support vector machine (LS-SVM) as an alternative multivariate calibration method for the simultaneous quantification of some common adulterants (starch, whey or sucrose) found in powdered milk samples, using near-infrared spectroscopy with direct measurements by diffuse reflectance. Due to the spectral differences of the three adulterants a nonlinear behavior is present when all groups of adulterants are in the same data set, making the use of linear methods such as partial least squares regression (PLSR) difficult. Excellent models were built using LS-SVM, with low prediction errors and superior performance in relation to PLSR. These results show it possible to built robust models to quantify some common adulterants in powdered milk using near-infrared spectroscopy and LS-SVM as a nonlinear multivariate calibration procedure.