Search

Introduction Metabolic syndrome (MetS), defined as a cluster of cardiometabolic factors, is a public health challenge because of its growing prevalence. In the context of personalized medicine, new tools are necessary to bring additional knowledge about MetS etiology, better stratify populations and customise strategies for prevention. The objective of this study was to characterize the MetS phenotypic spectrum using complementary untargeted metabolomics platforms (HRMS, RMN). Technological and methodological innovation A case-control study was designed within the Quebec NuAge cohort1. Six complementary untargeted metabolomic/lipidomic approaches were performed on serum samples collected at recruitment and 3 years later. Procedures were set up to guaranty the inter-laboratory standardisation from sample preparation to data processing, performed using reproducible online Galaxy workflows. A full feature selection strategy was developed to build a comprehensive molecular MetS signature, stable over time. Results and impact A wide range of metabolites (lipids, carbohydrates, amino-acids, peptides…) reflecting subject stability and providing new insights about underlying mechanisms, were found to be modulated. An optimized reduced signature was proposed, allowing good prediction performances (12% misclassification, AUC=0.95, CI:[0.92-0.98]). These results demonstrated the interest of a multidimensional molecular phenotyping as part of the next generation of medicine tools in the frame of noncommunicable diseases. References [1] Gaudreau P et al., 2007. Rejuvenation Res.10(3):377-386.

Metabolic syndrome (MetS), defined as a cluster of cardio-metabolic factors including obesity, hypertension, dysglycemia, and dyslipidemia, and mostly affecting older adults, is now a public health challenge because of its growing prevalence. In the context of personalized medicine/nutrition, new tools are necessary to bring additional knowledge about MetS etiology, better stratify populations and customise strategies for prevention. The objective of this study was to investigate the integration of data from complementary untargeted metabolomics platforms (HRMS, RMN) and technologies to characterize the MetS phenotypic spectrum. A case-control study was designed within the Quebec NuAge cohort1. Six complementary untargeted metabolomic/lipidomic approaches, available within the MetaboHUB infrastructure2, were performed on serum samples collected at recruitment and 3 years later. Standard operating procedures were designed to guaranty the inter-laboratory standardisation from sample preparation to data processing. Data analyses were performed using reproducible online Galaxy workflows3. A full feature selection strategy was developed to build a comprehensive molecular MetS signature, stable over time. Consistent cross-sectional and longitudinal data were observed with a wide range of metabolites (lipids, carbohydrates, amino-acids, peptides…) reflecting subject stability regarding MetS, and providing new insights about underlying mechanisms. Correlation network analysis contributed to explore the links between the molecular signature and clinical parameters. Additionally, an optimized reduced signature was proposed, allowing good prediction performances (12% misclassification, AUC=0.96, CI:[0.94-0.98]), for future clinical application. These results demonstrated the interest of a multidimensional molecular phenotyping as part of the next generation of medicine tools in the frame of non-communicable diseases.

Metabolomics generates complex data that need dedicated workflows to extract the meaningful information. For biological interpretation, experts are mainly focusing on metabolites rather than on the redundant different analytical species. Moreover, the high degree of correlation in datasets is a constraint for the use of statistical methods. In this context, we developed a new tool to detect analytical correlation into datasets. The algorithm principle is to group features from the same analyte and to propose one single representative per group. The user can define grouping criteria with various options including correlation coefficient, retention time, mass defect information. The representative feature can be determined following four methods according to the analytical technology. The present tool was compared to one of the most commonly used free package proposing a grouping method: ‘CAMERA’, using its Galaxy version ‘CAMERA.annotate’ available in workflow4Metabolomics (W4M; http://workflow4metabolomics.org). To illustrate its functionalities, a published dataset available on W4M (Thevenot et al., 2015) was used as an example. Within the 3,120 ions of the dataset, the tool allowed creating 2,651 groups, meaning that 15% of ions are proposed to be filtered because of analytical redundancies. The proposed tool subdivided more than 20 groups of more than 10 ions into smaller ones corresponding to individual annotated metabolites, thus demonstrating the efficiency and relevance of the present approach. As a key element in metabolomics data analysis, the tool will be available via the web-based galaxy platform W4M with different output files for network vizualisation and for further data analysis within workflows.

Aging is a dynamic process depending on intrinsic and extrinsic factors and its evolution is a continuum of transitions, involving multifaceted processes at multiple levels. It is recognized that frailty and sarcopenia are shared by the major age-related diseases thus contributing to elderly morbidity and mortality. Pre-frailty is still not well understood but it has been associated with global imbalance in several physiological systems, including inflammation, and in nutrition. Due to the complex phenotypes and underlying pathophysiology, the need for robust and multidimensional biomarkers is essential to move toward more personalized care. The objective of the present study was to better characterize the complexity of pre-frailty phenotype using untargeted metabolomics, in order to identify specific biomarkers, and study their stability over time. The approach was based on the NU-AGE project (clinicaltrials.gov , NCT01754012) that regrouped 1,250 free-living elderly people (65-79 y.o., men and women), free of major diseases, recruited within five European centers. Half of the volunteers were randomly assigned to an intervention group (1-year Mediterranean type diet). Presence of frailty was assessed by the criteria proposed by Fried et al. (2001). In this study, a sub-cohort consisting in 212 subjects (pre-frail and non-frail) from the Italian and Polish centers were selected for untargeted serum metabolomics at T0 (baseline) and T1 (follow-up). Univariate statistical analyses were performed to identify discriminant metabolites regarding pre-frailty status. Predictive models were then built using linear logistic regression and ROC curve analyses were used to evaluate multivariate models. Metabolomics enabled to discriminate sub-phenotypes of pre-frailty both at the gender level and depending on the pre-frailty progression and reversibility. The best resulting models included four different metabolites for each gender. They showed very good prediction capacity with AUCs of 0.93 (95% CI = 0.87-1) and 0.94 (95% CI = 0.87-1) for men and women, respectively. Additionally, early and/or predictive markers of pre-frailty were identified for both genders and the gender specific models showed also good performance (three metabolites; AUC = 0.82; 95% CI = 0.72-0.93) for men and very good for women (three metabolites; AUC = 0.92; 95% CI = 0.86-0.99). These results open the door, through multivariate strategies, to a possibility of monitoring the disease progression over time at a very early stage.

Metabolomics is a powerful phenotyping tool in nutrition and health research, generating massive and complex data that need dedicated treatments to enrich our knowledge of biological systems. In particular, to deeper investigate relations between environmental factors, phenotypes and metabolism, discriminant statistical analyses performed separately on metabolomic datasets, are often complemented by associations with metadata (anthropometric, clinical, nutritional and physical activity data…). Another relevant strategy is to perform a multi-block partial least squares discriminant analysis (MBPLSDA) that simultaneously analyses data available from different sources, allowing determining the importance of variables and variable blocks in discriminating groups of subjects, taking into account data structure in thematic blocks. In order to propose a full open-source standalone tool, the present objective was to develop an R package allowing all steps of MBPLSDA analysis for the joint analysis of metabolomic and additional data. The tool was based on the mbpls function of the ade4 R package, enriched with different functionalities, including some dedicated to discriminant analysis. Provided indicators help to determine the optimal number of components, to check the MBPLSDA model validity, and to evaluate the variability of its parameters and predictions. To illustrate the potential of the proposed tool and the associated procedure, MBPLSDA was applied to a real case study involving metabolomics, nutritional and clinical data from a human cohort. The availability of the different functionalities in a single R package allowed optimizing parameters for an efficient joint analysis of metabolomics and epidemiological data to obtain new insights into multidimensional phenotypes. In particular, we highlighted the impact of filtering the metabolomic variables beforehand, and the relevance of a MBPLSDA approach in comparison to a standard PLS-discriminant analysis method.

Metabolomics generates massive and complex data. Redundant different analytical species and the high degree of correlation in datasets is a constraint for the use of data mining/statistical methods and interpretation. In this context, we developed a new tool to detect analytical correlation into datasets without confounding them with biological correlations. Based on several parameters, such as a similarity measure, retention time, and mass information from known isotopes, adducts, or fragments, the algorithm principle is used to group features coming from the same analyte, and to propose one single representative per group. To illustrate the functionalities and added-value of this tool, it was applied to published datasets and compared to one of the most commonly used free packages proposing a grouping method for metabolomics data: 'CAMERA'. This tool was developed to be included in Galaxy and will be available in Workflow4Metabolomics (http://workflow4metabolomics.org). Source code is freely available for download under CeCILL 2.1 license at https://services.pfem.clermont.inra.fr/gitlab/grandpa /tool-acf and implement in Perl.

Context: Ageing is a dynamic process depending on intrinsic and extrinsic factors and its evolution is a continuum of transitions, involving multifaceted processes at multiple levels. It is recognized that frailty and sarcopenia are shared by the major age-related diseases thus contributing to elderly morbidity and mortality. They are major health issues in aging populations, given their high prevalence and association with several adverse outcomes. Pre-frailty is still not well understood but it has been associated with changes in several physiological systems, including inflammation as well as changes in the balance of micronutrients and vitamins. Due to the complex phenotypes and underlying pathophysiology, the need for robust and multidimensional biomarkers is now essential to move towards more personalized care and prevention. Objective: The objective of the present study was to better characterize the complexity of pre-frailty phenotype using untargeted metabolomics in order to identify specific biomarkers, and study their stability over time. Research design and methods: The approach was based on the NU-AGE project (FP7 EU programme; clinicaltrials.gov, NCT01754012) that regrouped 1,250 free-living elderly people (65-79 y.o., men and women), free of major diseases, recruited within five European centres. Half of the volunteers were randomly assigned to an intervention group (1-year Mediterranean type diet). Presence of frailty was assessed by the criteria proposed by Fried et al (Fried et al. J Gerontol A Biol Sci Med Sci, 2001). In this study, a sub-cohort consisting in 212 subjects (pre-frail and non-frail) from the Italian and Polish centres were selected for mass spectrometry-based untargeted metabolomics. Metabolic profiles were determined from serum samples at T0 (baseline) and T1 (follow-up). All data were processed under the Galaxy web-based platform Worflow4metabolomics, guaranteeing their reproducibility (Giacomoni et al. Bioinformatics, 2015). Univariate statistical analyses were performed to identify discriminant metabolites regarding pre-frailty status. Predictive models were then built using linear logistic regression and ROC curve analyses were used to evaluate multivariate biomarkers. Results: Presence of sub-phenotypes of pre-frailty both at the gender level and depending on the pre-frailty progression and reversibility were revealed by untargeted metabolomics. Additionally, early markers, able to predict the evolution towards pre-frailty within one year, were identified for both genders. Moreover, some of these early biomarkers were found to be still relevant for classification of a ‘light pre-frail’ phenotype after its clinical appearance. Conclusion: These results open the door, through multivariate strategies, to the possibility of monitoring the disease progression over time at a very early stage. Longitudinal analysis of individual time trajectories to detect early deviations of health status would indeed contribute to a better disease prevention.

Les approches métabolomiques non ciblées par GC-MS ont pu évoluer grâce au développement de nouveaux instruments de plus haute résolution, comme le GC-QToF. Cette approche est utilisée au laboratoire dans le cadre de projets scientifiques pour la recherche de biomarqueurs permettant la carac- térisation de phénotypes métaboliques. Une méthode d’analyse non ciblée pour la détermination de profils métaboliques de biofluides par GC- QToF a été adaptée d’une méthode basse résolution (Gao et al (1)) basée sur une double dérivation oximation /silylation. Cette technique, plus sensible et plus résolutive, nécessite des outils de traitement des données spéci- fiques dédiés. Aussi, nous avons dû adapter des outils développés par notre laboratoire pour le traitement de données métabolomiques à ce type de données. Ces outils comprennent l’extraction des données par xcms sous la plateforme Galaxy (W4M, (2)), ainsi que tout le workflow conduisant à l’annotation des ions extraits après filtration et correction des effets batch. Parallèlement, nous déployons une stratégie de déconvolution, à partir d’outils constructeurs afin de com- pléter les résultats obtenus sous Galaxy. A ce jour, les bibliothèques GC-MS (NIST, Golm, Massbank) restent très utilisées pour l’identification des métabolites mais ne contiennent aucun spectre avec des masses précises bien que certains provi- ennent de GC-EI-ToF. Par conséquent, nous constituons une bibliothèque interne en haute résolution avec des standards purs et en matrices biologiques qui alimentera la base de données PeakForest de l’infrastructure française MetaboHUB. La mesure des masses précises ainsi que le développement de nouveaux outils d’automatisation du traite- ment de données devraient permettre de lever certains verrous rencontrés dans la recherche de biomar- queurs concernant l’identification des métabolites.

Metabolomics is a powerful tool in nutrition research for characterizing complex phenotypes and identifying biomarkers of specific physiological responses to different diets in the context of molecular pathways, physiology and health status. However, nutritional metabolomics is particularly challenging because of the low amplitude of effects induced by nutritional interventions or transitions, in comparison to the large variability in samples due to intra/inter-individual variations but also collection, storage and pre-treatment. Untargeted approaches using multiple analytical platforms are particularly of relevance to obtain a large analytical coverage of metabolites, which is essential because of the extreme chemical diversity of the endogenous and food metabolomes, ranging from sugars and lipids to polyphenols. However, in this context, important criteria for selection of an untargeted method are not only based on analytical coverage, but also on robustness of sample pre-treatment and on analytical reproducibility in terms of retention times, masses and signal intensities. We will illustrate the use of UPLC/HRMS and GC/MS based untargeted strategies for the study of food exposure, as well as for the characterization of metabolic phenotypes associated with development or prevention of chronic metabolic diseases. Indeed, the determination and the annotation of plasma and urine metabolic profiles constitute a powerful data-driven approach for the discovery of potential biomarkers of food intake and/or metabolic changes. However, irrespective of the type of analytical techniques used, these metabolomic strategies generate large amounts of data that need analyses and integration to extract the biologically meaningful information and enable the discovery of new knowledge. We will show that application of numerical and symbolic data mining methods based on clustering and pattern mining/formal concept analysis are particularly relevant to discover classes and sub-classes of individual phenotypes, as well as guide visualization and interpretation of data from nutritional metabolomics