LEAPS-MPS: Optimal Design of Therapeutic Phage Cocktails: a Data-Driven Mathematical Approach

抗菌素耐药性已被描述为21世纪对人类健康的最大威胁之一。在人、动物、植物和环境中发现的耐抗生素细菌已经迅速出现并蔓延到世界各地。噬菌体，简称噬菌体，是一种进化为感染和杀死细菌的病毒。在抗生素广泛使用之前，噬菌体疗法在20世纪20年代和30年代成功地应用于治疗各种感染。作为一种已有百年历史的感染治疗方法，噬菌体疗法目前被视为一种潜在的抗生素替代品，并正在广泛地重新开发以治疗多药耐药感染。存在着数百万种噬菌体，每种噬菌体都具有不同的特性，因此不可能通过实验测试它们对单个临床病原体的功效。数学模型可以帮助识别一些特征，这些特征表明噬菌体是一种有希望的治疗候选者。该项目将引领噬菌体鸡尾酒和抗生素治疗最佳组合的发展，并进一步降低抗生素耐药性在人类-动物-植物-生态系统界面造成的健康风险。该项目将吸引具有不同研究背景和兴趣的本科生，并支持伍斯特学院STEM领域代表性不足的学生。学生将受益于全年的跨学科培训，将真正的生命科学问题转化为标准的数学问题。该项目将进一步推进课程和项目的发展，这将增强该机构的研究环境，并进一步建立一个持续的、以学生为中心的跨学科数学生物学研究项目。该项目将通过使用非线性常微分方程来模拟多噬菌体与细菌之间的密度依赖相互作用，探索三种假单胞菌噬菌体的单、双同时和双顺序给药策略的治疗效果。不同于以往的研究集中在噬菌体治疗的噬菌体杀伤方面，本项目将结合噬菌体杀伤和细菌噬菌体耐药性的进化。了解这两者对于噬菌体鸡尾酒组合和抗生素治疗的未来发展至关重要。由于吸附速率与细菌生长有关，研究者将考虑比通常使用的线性相互作用更具有生物学真实性的混合，研究随机建模方法以捕获噬菌体抗性细菌的出现，并使用分岔和敏感性分析分析模型的结构，为进一步的建模修改提供信息。该模型框架在临床研究中具有许多推理和探索性的用途，如确定与噬菌体特征相对应的最敏感的模型参数，进行噬菌体选择，以及探索不同的治疗方案。物种特异性和噬菌体特异性建模研究可以作为预测分析工具，帮助设计未来的临床研究，并提高对实验数据的理解。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Antimicrobial resistance has been described as one of the biggest threats to human health in the twenty-first century. Antibiotic-resistant bacteria, found in people, animals, plants, and the environment, have rapidly emerged and spread throughout the world. Bacteriophages, or phages for short, are viruses that have evolved to infect and kill bacteria. Before the widespread use of antibiotics, phage therapy was successfully applied for the treatment of a variety of infections in the 1920s and 1930s. As a century-old infection remedy, phage therapy is currently viewed as a potential antibiotic alternative and is being widely redeveloped to treat multidrug-resistance infections. There are millions of phages in existence, each with different properties, it is therefore impossible to experimentally test them for efficacy against individual clinical pathogens. Mathematical models can help identify characteristics that would suggest that a phage is a promising therapeutic candidate. This project will lead to the development of an optimal combination phage cocktail and antibiotic therapy and further reduce the health risks at the human-animal-plant-ecosystem interface caused by antimicrobial resistance. The project will engage undergraduate students with different research backgrounds and interests and support underrepresented students in STEM at The College of Wooster. Students will benefit from receiving year-round interdisciplinary training in formulating genuine life science questions into standard mathematical problems. The project will additionally advance curricular and program development, which will enhance the institution's research environment and further establish a sustained, student-focused, and interdisciplinary research program in mathematical biology at Wooster. The project will explore the treatment efficacies of single, double simultaneous, and double sequential administration strategies with three Pseudomonas phages by using nonlinear ordinary differential equations to model the density-dependent interactions between multiple-phage and bacteria. Unlike previous studies that focused on the phage-killing aspect of phage therapy, this project will incorporate both phage-killing and the evolution of bacterial phage resistance. Understanding both is important for the future development of a combination phage cocktail and antibiotic therapy. The investigator will consider a more biologically realistic mixing than the commonly used linear interaction, as adsorption rate is linked to bacterial growth, investigate the stochastic modeling approach to capture the emergence of phage-resistant bacteria, and analyze the structure of the model using bifurcation and sensitivity analyses to inform further modeling modifications. The modeling framework has many inferential and exploratory uses for clinical investigation such as identifying the most sensitive model parameters, corresponding to phage characteristics, for phage selection as well as exploring different treatment regimens. The species-specific and phage-specific modeling studies can be used as predictive analytics tools to assist in the design of future clinical studies and to improve the understanding of experimental data.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.