On the Use of Correlation and MI as a Measure of Metabolite-Metabolite Association for Network Differential Connectivity Analysis.

Sanjeevan Jahagirdar,Edoardo Saccenti

doi:10.3390/metabo10040171

Abstract

Metabolite differential connectivity analysis has been successful in investigating potential molecular mechanisms underlying different conditions in biological systems. Correlation and Mutual Information (MI) are two of the most common measures to quantify association and for building metabolite—metabolite association networks and to calculate differential connectivity. In this study, we investigated the performance of correlation and MI to identify significantly differentially connected metabolites. These association measures were compared on (i) 23 publicly available metabolomic data sets and 7 data sets from other fields, (ii) simulated data with known correlation structures, and (iii) data generated using a dynamic metabolic model to simulate real-life observed metabolite concentration profiles. In all cases, we found more differentially connected metabolites when using correlation indices as a measure for association than MI. We also observed that different MI estimation algorithms resulted in difference in performance when applied to data generated using a dynamic model. We concluded that there is no significant benefit in using MI as a replacement for standard Pearson’s or Spearman’s correlation when the application is to quantify and detect differentially connected metabolites.

Highlights

Metabolite concentration profiles measured in samples, like blood, urine, or tissues and their patterns of variations, are regulated by complex bio-molecular machines
Correlation and Mutual Information (MI) measures have been widely used in many research applications to quantify and describe the relationships between variables, having become the foundations for network inference methods [8]
Researchers trained in statistics tend to use correlation based indices, while researchers trained in computer science gravitate towards mutual-information

Summary

Introduction

Metabolite concentration profiles measured in samples, like blood, urine, or tissues and their patterns of variations, are regulated by complex bio-molecular machines. A biological system can be represented as a complex network of interconnected biomolecular entities [4] which can be visualised in a graphical manner as networks, i.e., sets of nodes that are connected by edges to indicate the existence and the strength of pairwise relationships [5]. This representation shifts the focus towards the relationships among biological entities rather than on their levels; in this light, network and network analysis are fundamental tools from the systems biology toolbox to investigate and understand metabolomic data [6]. When the nodes are metabolites, the network can be called a metabolite-metabolite association network [6,7], and, in modern metabolomic studies, the interest is to reconstruct these associations patterns from observed data measured in well designed experiments.

Methods

Results

Conclusion