Systems level analysis of animal metabolism by multicompartmental graph- and constraints-based modelling

One goal of systems biology is to explain and predict how an organism works in terms of underlying processes (metabolism, gene expression patterns etc.). The purpose of this application is to design and construct systems-level models of metabolism in an animal comprising multiple organs with overlapping but appreciably different metabolic repertoires. Our core system is the pea aphid Acyrthosiphon pisum, which bears obligate intracellular bacteria Buchnera aphidicola in a specialized organ, the bacteriome. The bacteria and bacteriome are treated as metabolic compartments that interact with the gut, fat body (the principal insect organ of intermediary metabolism) and embryos (accounting for up to 75% of the total biomass of aphids) via the insect circulatory system. Our modelling, thereby, explores the integration of metabolic networks across multiple components comprising two distinct genomes within a single organism. We particularly address two challenges: (1) To develop new algorithms leading to a better understanding of the structure of metabolic networks, improved routes to explore the networks systematically for extracting functional information, and a grasp of how such networks have arisen. (2) To devise multi-compartment models, using flux balance analysis, to explore the metabolic organisation of animals. The modeling process follows a step-wise increase in complexity and realism of metabolic integration, building from our previous metabolic reconstruction of Buchnera. The final model seeks to define the fully interactive metabolic network comprising Buchnera-aphid interactions in maternal tissues and embryos. In parallel, our models will be made available for use by the wider community.

From a metabolic perspective, an animal comprises a set of metabolic compartments (including organs) connected by the inter-compartment transfer of metabolites. We seek to construct and analyse multi-compartment models of metabolism, by transforming a single global metabolic network into a set of compartment networks of overlapping composition using flux balance analysis. In parallel, we will develop and apply two computational approaches for network analysis: an exploration of metabolic network motifs, and elementary modes to characterize and compare networks. Large inter-compartment metabolic differences facilitating model development will be achieved in our system comprising two distinct genomes within a single animal: the pea aphid Acyrthosiphon pisum and its symbiotic bacteria Buchnera housed in a special organ, the bacteriome. The first model comprises two compartments, describing how the insect and bacterial networks mediate the transformation of precisely known diet components to yield biomass production (and waste); and subsequent models are of increasing complexity, involving the subdivision of the insect network into overlapping, but distinct, networks of the contributing compartments/organs. Model construction will be informed by empirical data collected on the aphid transcriptome, Buchnera proteome and embryo nutrient budgets within this project, and the model outputs will be validated against whole-insect metabolic and nutritional datasets obtained previously from the investigators' labs. Algorithms and models developed in this project will be maintained in a central Repository linked to pre-existing databases, suitable for data-sharing within the project and dissemination to the wider community. The outputs will include metabolic network data for a range of insects with well-developed genomic resources.