Aggregation of Biochemical and Genetic Evidence Allows Refinement of Metabolic Models for Various Plant Tissues

The effective metabolic and genetic engineering of plants will be of great benefit to human society at many levels but is nevertheless confounded by the layers of complex biological networks that affect plant growth and the production of key metabolites under different conditions. The modeling community has responded to this challenge by publishing several large metabolic networks that can be manipulated and explored in silico in order to benefit research efforts in vivo. However, all hypotheses that a researcher may explore in such metabolic reconstructions are dependent on the gene-reaction associations. Over-annotation in public biochemical databases is common and leads to an inflated number of gene-reaction associations and unreliable predictions of metabolic phenotypes. We introduce a process of building metabolic reconstructions based on the biochemical and genetic evidence that is available, demonstrating significant improvements in the number of the gene-reaction associations in a genome-scale model for maize. Furthermore, we present a new approach for integrating transcript profiles with the genome-scale model, allowing us to construct several refined tissue-specific metabolic models for the maize leaf, embryo, and endosperm. We validate our models using fluxomics data for the endosperm and embryo, demonstrating an improved capacity of our models to fit the available fluxomics data. All models and methods are publicly available via the DOE Systems Biology Knowledgebase, paving the way for further in silico studies with a wide variety of plant genomes.