Chemometric approaches to improve PLSDA model outcome for predicting human non-alcoholic fatty liver disease using UPLC-MS as a metabolic profiling tool

Abstract

An MS-based metabolomics strategy including variable selection and PLSDA analysis has been assessed as a tool to discriminate between non-steatotic and steatotic human liver profiles. Different chemometric approaches for uninformative variable elimination were performed by using two of the most common software packages employed in the field of metabolomics (i.e., MATLAB and SIMCA-P). The first considered approach was performed with MATLAB where the PLS regression vector coefficient values were used to classify variables as informative or not. The second approach was run under SIMCA-P, where variable selection was performed according to both the PLS regression vector coefficients and VIP scores. PLSDA models performance features, such as model validation, variable selection criteria, and potential biomarker output, were assessed for comparison purposes. One interesting finding is that variable selection improved the classification predictiveness of all the models by facilitating metabolite identification and providing enhanced insight into the metabolic information acquired by the UPLC-MS method. The results prove that the proposed strategy is a potentially straightforward approach to improve model performance. Among others, GSH, lysophospholipids and bile acids were found to be the most important altered metabolites in the metabolomic profiles studied. However, further research and more in-depth biochemical interpretations are needed to unambiguously propose them as disease biomarkers.

Metabolomic Changes in the Rat Retina After Optic Nerve Crush

“…Further details of sample acquisition and data preprocessing can be found in a previously published procedure. 45 A principal component analysis (PCA) was done to achieve the natural interrelationship (grouping, clustering, or outlier detection) among samples and quality controls. Supervised multivariate data analysis was performed by partial least-squares discriminant analysis (PLSDA).…”

“…Two PLSDA models were calculated after performing a variable importance in the projection (VIP) selection procedure. 45 Model validation was performed by typical 7-fold CV, the R 2 values indicated the goodness of the model and the Q 2 values estimated the model's predictive ability. These values were used to assess the performance of the model and to select the optimal number of principal components (t).…”

Validación y selección de indicadores para la evaluación de competencias. Aplicación a la competencia transversal de creatividad, innovación y emprendimiento

“…A partir de la utilización de modelos PLS, se ha llevado a cabo la selección de indicadores mediante la integración de un proceso de selección basado en distribuciones libres, en una secuencia de doble validación cruzada (Quintás et al, 2012). Dicha técnica consiste en los siguientes pasos: Determinación del poder de predicción con el Validation set.…”

Chemometric Approaches for Systems Biology

“…Hence, the goal is to perform a variable selection on the third mode of the data, the spectral bands, to assess whether a few wavelengths (three to five) 161 have enough discriminant power to classify each fruit correctly. Permutation testing is used since it is one of the most used techniques to perform variable selection in PLS-DA [213,222,223,298].…”

“…Especially PLS discriminant analysis (PLS-DA) is commonly applied to distinguish between biological conditions, such as a particular illness. For example, in [213] this technique was used to discriminate between non-steatotic and steatotic human liver profiles, and in [214] PLS-DA was used for the diagnosis of inherited metabolic disorders (IMDs), analysing plasma and blood samples of subjects with phenylketonuria and medium chain acyl CoA dehydrogenase deficiency. In these studies, the PLS-DA model is used for finding potential biomarkers among the pool of metabolites analised.…”