Every minute, each one of our trillions of cells carries out countless chemical reactions to convert food into energy, build proteins and eliminate toxic waste for instance. Taken as a whole, all these reactions make up our metabolism. But by which underlying mechanisms is it regulated? Researchers from SIB and the CHUV (Lausanne University Hospital) have found a new method to understand the regulatory basis of metabolism at a far higher resolution.


1. What are genome-scale metabolic networks (GSMN)?
GSMN are complex biological networks that interconnect thousands of nodes themselves divided into four main types: 1) metabolites; 2) biochemical and transport reactions; 3) enzymes and transporters; and 4) genes. Provided sufficient data is at hand, GSMN can be used as a predictive tool to understand whether and how an organism produces a certain amount of biomass, or how essential a gene is for an organism by simulating knock-out experiments.
2. What is transcriptomics?
The science of transcriptomics examines the expression level of RNAs in a given cell population, thereby revealing which genes are actively expressed at any given point in time. One of the most popular methods for studying the transcriptome and for measuring RNA levels is to use next-generation sequencing technologies, called RNAseq.

Being able to pinpoint the regulatory pathways that underly our metabolism would help us better understand – and perhaps even prevent – potential malfunctions, such as those related to diabetes or obesity. Teasing these pathways apart however, requires not only a high-level understanding of the biological and chemical processes at stake such as that provided by genome scale metabolic networks (GSMN, see box 1), but also a fine resolution of what each gene does as provided by expression – or transcriptomics – data (RNAseq, see box 2).

With researchers from the CHUV, SIB Scientists from the Vital-IT Group have developed a new method that combines both approaches, thus offering a finer understanding of our metabolism’s underlying gene regulations as well as the means to predict them. Their method is available as an R package called metaboGSE, and the results are published in the journal Bioinformatics.

“By making use of already available data, this “combined” method enables researchers to obtain metabolic information at a greater level of specificity than they would by using GSMN or RNAseq approaches separately,” says SIB’s senior scientist and author Marco Pagni. “In a way, it’s like looking through a magnifying glass, and instead of seeing ‘phospholipid biosynthesis, we see ‘phosphatidylcholine biosynthesis’ – which, in terms of resolution, is a big step forward”.

The team is calling for a broad reexamination of metabolic studies for which good quality RNAseq data and GSMN are already available, in order to acquire more robust and predictive results.

Tested and validated on mouse adipocyte data related to obesity traits, the team now intends to apply the same method on human diabetes-related data from several European cohorts. This would be in the framework of several Innovative Medicines Initiative (IMI) projects, where SIB is acting as the Data Coordination Centre (read more).



Tran V D T et al. Condition-specific series of metabolic sub-networks and its application for gene set enrichment analysis. Bioinformatics doi:10.1093/bioinformatics/bty929

Illustration of the autophagy process by David S. Goodsell, the Scripps Research Institute (http://pdb101.rcsb.org/motm/203)