Mapping Microbial Community Structure and Function in Enhanced Biological Phosphorus Systems

增强型生物除磷（EBPR）是活性污泥技术的一种变体，除了使用微生物联合体去除有机物外，还可以从进水废水中去除磷，以防止接收体的富营养化。与活性污泥一样，采用EBPR的废水处理厂由于功能微生物的种群失衡而表现不佳。迄今为止，对性能恶化背后的根本原因以及可靠的纠正措施的了解都是难以捉摸的。非培养微生物学技术和宏基因组学的进步促进了对这些系统中微生物群落的多样性、代表性个体门的相对丰度以及各自功能生态学的原位特征的研究。补充这项工作的是实验室规模的研究，利用高度富集的细菌培养物，以及对全尺寸设施的调查，这些设施将感兴趣的微生物群落核心成员的相对丰度与各自功能的理化分析相关联。然而，就目前的形式而言，现有资料的性质仍然是支离破碎的，而且由于这些资料是在孤立的情况下产生的，不具有结论性——更不用说在理解所报告的技术不稳定性方面是相互矛盾的。鉴于此，本博士项目的总体目标是开发和部署统计模型，用于评估和统一（1a）微生物群落组成，（2a）各自的生态功能和门间相互作用以及（3a）工艺设计和操作的要素。因此，该项目的组成目标包括但不限于：（1b）收集现有的信息库并将其纳入统一的工作数据库；（2b）构建一个网络模型，用于表示、链接和评估组装的数据；（3b）调查和评估相关的统计工具，如自学习贝叶斯统计模型，以便利用现有证据推断EBPR微生物群落中不同群体之间关系的性质，以及它们与工艺配置、操作参数和环境条件的关系。该项目的目标是提高对与EBPR相关的微生物群落的整体理解。特别是，该项目的潜在影响将（1c）能够制定具体的战略，以培养合适的社区，确保使用该技术的废水处理设施始终具有令人满意的性能，从而达到合规性；（2c）为未来EBPR过程的设计优化提供框架；（3c）提供工具，以便系统地识别知识缺陷。从而提供了未来关键研究领域的路线图。从更广泛的意义上讲，本项目拟开发的工具可应用于任何生物系统的开发，可回答以下问题：（1d）在全面治疗和/或生产设施中监测的关键参数，以确保有效的性能，（2d）该数据池所需的（最小）样本量是多少，（3d）如何利用现有数据，或（4d）如何定义众包策略以更好地做出明智的决策。该项目将把水工程领域与统计和应用概率领域联系起来，这两个领域是EPSRC主题研究组合指出的关键增长领域之一。

Enhanced Biological Phosphorus Removal (EBPR) is a variation of the activated sludge technology that in addition to removal of organic matter using a microbial consortium, enacts phosphorus removal from the influent wastewater to prevent eutrophication of the receiving body. Like activated sludge, wastewater treatment plants employing EBPR are known to suffer from poor performance attributed to population imbalances of the functional microorganisms. An understanding of the root causes behind the deteriorated performance, and of the reliable corrective measures have thus far proven themselves elusive.Advances in culture-independent microbiological techniques and metagenomics have stimulated work towards the in situ characterisation of microbial communities in these systems in terms of their diversity, relative abundance of the representative individual phyla, as well as their respective functional ecology. Complementing this work are laboratory-scale studies utilising highly enriched bacterial cultures, as well as surveys of full-scale facilities which correlate the relative abundance of perceived core members of the microbial community of interest in tandem with physiochemical analyses to their respective function. In its current form however, the nature of the available information remains fragmented and, owing to the isolation in which it was produced, inconclusive - not to mention that with regards to understanding the reported instability of the technology, contradictory.In light of this, the overall aim of this PhD project is to develop and deploy statistical models with which to evaluate and unite elements of the (1a) composition of the microbial community, (2a) their respective ecological function and inter-phyla interactions and (3a) process design and operation.The component objectives of this project thus include but are not limited to: (1b) collecting and framing the existing pool of information into a unified working database; (2b) constructing a network model with which to represent, link and evaluate the assembled data; and (3b) investigating and assessing relevant statistical tools such as a self-learning Bayesian statistical models in order to make use of the available evidence to infer the nature of the relationships among different groups within EBPR microbial consortia, their relation to the process configuration and operational parameters and environmental conditions.The ambition of this project is to improve the holistic understanding of the microbial consortia relevant to EBPR. In particular, the potential impact of this project will (1c) enable the formulation of concrete strategies to cultivate the right communities that would ensure consistently satisfactory performance, and therefore compliance, of wastewater treatment facilities employing this technology, (2c) provide the framework for design optimisation of future EBPR processes and (3c) provide the tools to enable systematic identification of knowledge deficiencies, thereby providing a roadmap of key areas of future research.In a broader sense, the proposed tools to be developed in this project could find application in the exploitation of any biological system, answering questions such as: (1d) which are the key parameters to monitor in full-scale treatment and/or production facilities to ensure efficient performance, (2d) what is the (smallest) required sample size of this pool of data, (3d) how can one make use of the data already available, or (4d) how to define crowd-sourcing strategies for better informed decision making.This project will connect the field of water engineering to that of statistics & applied probability, two among the key growth areas as indicated by the EPSRC thematic research portfolio.