Community-scale models of microbiomes: Articulating metabolic modelling and metagenome sequencing

Abstract

Building models is essential for understanding the functions and dynamics of microbial communities. Metabolic models built on genome-scale metabolic network reconstructions (GENREs) are especially relevant as a means to decipher the complex interactions occurring among species. Model reconstruction increasingly relies on metagenomics, which permits direct characterisation of naturally occurring communities that may contain organisms that cannot be isolated or cultured. In this review, we provide an overview of the field of metabolic modelling and its increasing reliance on and synergy with metagenomics and bioinformatics. We survey the means of assigning functions and reconstructing metabolic networks from (meta-)genomes, and present the variety and mathematical fundamentals of metabolic models that foster the understanding of microbial dynamics. We emphasise the characterisation of interactions and the scaling of model construction to large communities, two important bottlenecks in the applicability of these models. We give an overview of the current state of the art in metagenome sequencing and bioinformatics analysis, focusing on the reconstruction of genomes in microbial communities. Metagenomics benefits tremendously from third-generation sequencing, and we discuss the opportunities of long-read sequencing, strain-level characterisation and eukaryotic metagenomics. We aim at providing algorithmic and mathematical support, together with tool and application resources, that permit bridging the gap between metagenomics and metabolic modelling.

FIGURE 1

From mono‐organism to community function‐based model of microbial populations. Monoculture model (upper panel): in monoculture, a GENRE is obtained from the annotated genome, with semi‐automatic tools (e.g. gap‐filling) and manual curation. A matrix representation of the metabolic network, the reaction matrix, is an input for flux prediction models. Metabolic fluxes can be computed iteratively to estimate changes over time of a dynamical model of the population. Community model (lower panel): in a microbial community, the metagenome makes it possible to obtain MAGs and GENREs for members of the community. GENREs may share some common functions and metabolites (vertical grey vertices in the GENRE batch). A first paradigm for building a community‐wide metabolic model is a bag‐of‐genome model: A set of independent GENREs interacting through common external metabolites. The corresponding reaction matrix is the concatenation of the individual matrices, and the corresponding dynamical model is the sum of the fluxes computed for each unique genome. A second paradigm termed bag‐of‐gene model consists in merging all the GENREs through their common functions and metabolites, resulting in a global metabolic network for the whole community. The corresponding reaction matrix is the concatenation of the individual matrices after merging or removal of common reactions (columns) and metabolites (rows), resulting in a lower dimension matrix and a dynamical system with a unique equation. With this paradigm, analysis can be performed at the scale of a community, but the contribution of each taxon is fuzzier.

FIGURE 2

Approaches for numerical simulation and reasoning on community metabolism. Methods and tools are sorted according to the functional characterisation of the microbial community metabolism they can provide: predicting small‐scale community metabolism (upper left panel), assessing community dynamics (lower left panel), deciphering cooperation and competition potentials (upper right panel) and characterising large‐scale community metabolism (lower right panel). For each class, the main methods and dedicated software (grey italic names) are recapitulated. The main ‘pros’ (+) and ‘cons’ (–) are listed on the lateral box associated with each panel.

FIGURE 3

Dynamic models of metabolism: opportunities beyond GENRE‐based models. Different formalisms can be used to capture microbial population dynamics (upper panel). These models can be placed on a gradient, from phenomenological (left) to mechanistic models (right) of the metabolic capabilities of the community. Each formalism can be supplemented by additional spatio‐temporal mechanisms (lower panel) that can be represented at the population (PDE) or individual scale (IBM). The addition of spatio‐temporal features in the community dynamics induces simplifications in the metabolic models to keep the model computationally tractable.

FIGURE 4

Metagenomic sequencing and data processing. After quality control, shotgun metagenomic reads can enter an assembly pipeline or an assembly‐independent profiling pipeline. The assembly pipeline is highlighted as it permits the reference‐free reconstruction of MAGs and subsequently GENRE reconstruction that directly survey the functions present in the ecosystem. Examples of tools associated with each step are indicated next to the box. Tools or steps related specifically to TGS, eukaryotic metagenomics and strain resolution are highlighted with dotted boxes.