From correlation to causation: analysis of metabolomics data using systems biology approaches

Abstract

Introduction: Metabolomics is a well-established tool in systems biology, especially in the top-down approach. Metabolomics experiments often results in discovery studies that provide intriguing biological hypotheses but rarely offer mechanistic explanation of such findings. In this light, the interpretation of metabolomics data can be boosted by deploying systems biology approaches.

Objectives: This review aims to provide an overview of systems biology approaches that are relevant to metabolomics and to discuss some successful applications of these methods.

Methods: We review the most recent applications of systems biology tools in the field of metabolomics, such as network inference and analysis, metabolic modelling and pathways analysis.

Results: We offer an ample overview of systems biology tools that can be applied to address metabolomics problems. The characteristics and application results of these tools are discussed also in a comparative manner.

Conclusions: Systems biology-enhanced analysis of metabolomics data can provide insights into the molecular mechanisms originating the observed metabolic profiles and enhance the scientific impact of metabolomics studies.

Keywords: Association network; Correlation network; Enrichment analysis; Network analysis; Pathway.

Fig. 1

Relationship between the systems biology cycle and the metabolomics pipeline

Fig. 2

Association network of 133 blood metabolites measured using MS/MS on 2139 subjects. a Plasma metabolites association networks obtained using the four different methods. b Serum metabolites association networks obtained using the four different methods. c Consensus association network for serum and plasma. CLR context likelihood of relatedness, ARACNE algorithm for the reconstruction of accurate cellular networks, PCLRC probabilistic context likelihood of relatedness on correlations, CORR Pearson’s correlation). Reproduced with permission from Suarez-Diez et al. (2017). Copyright (2017) American Chemical Society

Fig. 3

a Weight plot and b loadings plot of the INDSCAL model for the metabolite correlation network obtained using the PCLCR method. Each dot represents a network that corresponds to a given cardiovascular (CVD) risk parameter. Blue dots indicate low latent CVD risk, while red indicate high latent CVD risk. The associated CVD risk parameters are indicated in upper case for high risk and lower case for low risk. A reference network (indicated as “All”, black ball), built using all the subjects in the study, is given as reference. Reproduced with permission from Saccenti et al. (2014). Copyright (2014) American Chemical Society

Fig. 4

Overview of metabolic flux modelling using stable isotope resolved metabolomics data

Fig. 5

Overview of the Global test. a From the autoscaled data matrix, m metabolites belonging to the same pathway are selected. A binary outcome is defined, coded 0 and 1, for instance healthy versus disease. b A score statistic Q is calculated from the mean centered outcome and the matrix of selected metabolites. c The significance of the relation between the group of metabolites (pathway) and the outcome is determined by performing a permutation test. Reproduced with permission from Hendrickx et al. (2012); Copyright (2012) Elsevier B. V