Metabolic networks have scale free topologies, quite different from random networks and suggestive of underlying rules that govern their assembly and expansion. While many insights have emerged through theoretical and computational modeling of metabolism, few studies have focused explicitly on analyzing the evolution of metabolic communities, particularly those found in extreme environments. In this study we combine network analysis of metagenomic data sets from environments such as Yellow Stone National Park with the development of novel computational models to investigate how metabolic communities evolve in response to catalytic innovation. The specific objectives of this project are to (1) Build computational models to investigate network evolution of metabolic communities. (2) Identify the network architecture and patterns of diversity for metagenomes, and (3) Utilize computational models as a predictive framework for testing novel hypothesis about the structure and evolution of metabolic communities, particularly in extreme environments. Our methods include building computational models of multilevel evolution of chemical reaction networks using kinetic Monte Carlo algorithms with codes developed in Python. Results will be compared to networks inferred from 1) whole genomes and 2) metagenomes of complex microbial communities, which will be analyzed using techniques from network and graph theory. Our proposal is directly relevant to exobiology. In particular, our research proposal addresses the first of the three basic questions emphasized in the NASA Astrobiology Roadmap: 'how does life begin and evolve?' Our project will directly contribute to four of the roadmap objectives identified under three of its seven goals. This includes Goal 4 to 'Understand how life on Earth and its planetary environment have co-evolved through geological time'. In particular, we will address Objective 4.2 aimed at understanding the 'Production of complex life' by investigating the pathways by which metabolic networks expand and evolve to permit increasingly complex ecosystems. We will also address Goal 5, to 'Understand the evolutionary mechanisms and environmental limits of life' with this project. By characterizing the evolution of the network structure and diversity of metabolic communities (including those in extreme environments such as Yellow Stone National Park) our proposed research will specifically contribute to Objective 5.2, addressing the 'Co-evolution of microbial communities' and Objective 5.3, addressing 'Biochemical adaptation to extreme environments'. Finally, our project furthers Goal 6 to 'Understand the principles that will shape the future of life, both on Earth and beyond' by answering the call in Objective 6.1 to address the 'Effects of environmental changes on microbial ecosystems'. We do so by developing predictive models that integrate biogeochemistry with adaptation by microbial ecosystems through our utilization of environmental metagenomic data sets in the construction of our computational model systems. More specific to this funding opportunity, our project precisely conforms to the NASA Exobiology area of emphasis in 'Early Evolution of Life and the Biosphere'. Our project will directly contribute to objective (iv) in this opportunity to ``study the coevolution of microbial communities, and the interactions within such communities, that drive major geochemical cycles, including the processes through which new species are added to extant communities'' by detailing the mechanisms underlying the expansion of chemical networks in metabolic communities. The results of our study will also be relevant to 'Biosignatures and Life Elsewhere' under by providing a novel computational framework for understanding the multilevel evolution of metabolic reaction networks that will ``constrain or extend concepts of possible chemical evolution relevant to the origin, evolution, and distribution of life.''