An Agent-Based Modeling Platform for Environmental Biotechnology

环境中的有害污染物继续威胁着公众健康和环境 安全为代价的人类暴露于主要污染物类别，如多氟化合物 （PFCs），有害有机化合物（HOCs）和重金属，已被链接到各种 并遵守严格的州和联邦环境法规。 生物修复是一种低成本、环境友好的方法， 然而，传统的生物修复技术可能会遭受不可靠性，低 降解速率和不完全降解。作为超级基金网站和其他网站的利益相关者 随着水或土壤污染的迫切需要更有效，更低成本和更可靠 补救技术，关键是要看在计算上的进步 建模以开发下一代精密工程生物修复技术。 拟议的项目建立在第一阶段的成功成果的基础上，在第一阶段，一个新的计算 平台的设计和验证，以准确地预测生物修复动力学 一个多生物体的微观世界，降解地下水中的有机氯化合物。的基础 该平台是一种称为基于代理的建模（ABM）的方法， 复杂生态系统中的单个组分（例如微生物）用于预测和 优化系统级性能（例如生物修复动力学）。 在第二阶段项目中，第一阶段开发的新型计算平台是 通过利用生物信息学的机器学习组件进一步改进 数据库，开发合理定制的微生物组，用于降解复杂污染物 混合物。迭代实验验证模型输出进行了创新的 材料科学平台，保持不同物种的相对集中， 多区处理屏障（原位）或多区生物反应器内的微生物组恒定 （异地）。该项目包括重点开发一个生物修复用例的原型， 其直接与常规（非精确）生物修复系统处理相比， 地下水实际污染。这将是为了评估和量化 利用该项目的新的计算技术的预期技术和经济效益 生物技术发展平台。 拟议项目的广泛，长期影响将是改变发展和 通过整合计算建模、机器 学习、生物信息学和材料科学。通过利用跨学科的新颖工具， 该项目将加速开发更精确，可靠和廉价的技术， 环境修复。该项目的成功还将提供新的 为工业界和学术界提供合作机会，以更快地解决 环境中的高优先级污染物，并最终帮助减轻危险的影响， 污染物对受环境污染影响的社区的影响。

Hazardous pollutants in the environment continue to threaten public health and environmental safety. Human exposure to major contaminant classes, such as polyfluorinated compounds (PFCs), hazardous organic compounds (HOCs), and heavy metals, has been linked to a variety of diseases and is subject to stringent State and Federal environmental regulations. Bioremediation is a low-cost and environmentally friendly approach with many successful use-cases; however, conventional bioremediation technologies can suffer from unreliability, low degradation rates, and incomplete degradation. As stakeholders to Superfund sites and other sites with water or soil pollution urgently demand more efficient, less costly and more reliable remediation technologies, it is critical to look to advancements in computational modeling to develop next-generation, precision-engineered bioremediation technologies. The proposed project builds on successful outcomes from Phase I in which a new computational platform was designed and validated to accurately predict the bioremediation kinetics of a multi-organism microcosm degrading a combination of HOCs in groundwater. The basis of this platform is an approach called agent-based modeling (ABM), where the functions of individual components (e.g. microorganisms) within complex ecosystems are used to predict and optimize system-level properties (e.g. bioremediation kinetics). In this Phase II project, the novel computational platform developed in Phase I is further improved with a machine learning component that leverages bioinformatics databases to develop rationally tailored microbiomes for degrading complex pollutant mixtures. Iterative experimental validation of model outputs is conducted using an innovative materials science platform that maintains the relative concentration of different species in the microbiome constant within the multi-zone treatment barrier (in-situ) or multi-zone bioreactor (ex-situ). The project includes focused development of a prototype for one bioremediation use-case, which is directly compared to a conventional (non-precision) bioremediation system treating actual contaminated groundwater. This will be performed in order to assess and quantify the expected technical and economic benefits of harnessing the project's novel computational platform in biotechnology development. The broad, long-term impact of the proposed project will be to transform the development and implementation of bioremediation by integrating advancements in computational modeling, machine learning, bioinformatics, and materials science. By leveraging novel tools across disciplines, the project will accelerate the development of more precise, reliable and inexpensive technologies for environmental remediation. The successful outcome of the proposed project will also provide new collaborative opportunities for industry and academia to more rapidly address the remediation of high-priority pollutants in the environment, and ultimately help mitigate the effects of hazardous pollutants on communities impacted by the presence of environmental contamination.