Carbohydrate enzyme gene clusters in human gut microbiome

项目总结 人体肠道微生物组中的碳水化合物酶基因簇 我们的R01父项目(R01GM140370)打算开发四个用于自动标注的生物信息学工具 人类肠道微生物组中的CAZymes(碳水化合物活性酶)和CAZyme基因簇(CGCs)。 这些自动化工具将增强：(I)表征新多糖(或)的基础生物医学科学 糖)代谢酶和多糖利用位点(Puls，已知碳水化合物的基因簇 (2)新兴的个性化营养实践(例如，使用肠道 微生物组测序，以推断一个人是否对某些饮食多糖或益生元有反应)。 AI/ML应用需要两种类型的数据：(1)PULS(实验表征的基因簇 具有已知的碳水化合物底物)和(2)CGCs(不含已知的碳水化合物 底物)从人类微生物群中预测。 尽管父R01项目的重点是开发新的ML工具，但仍然存在挑战和 需要额外的支持才能为父项目中使用/产生的数据启用AI/ML就绪。 这些挑战包括：(I)训练数据大小很小，更多的PULS等待从文献中挑选出来 父R01项目不支持，(Ii)父R01项目不考虑制作 PULS和CGCS AI/ML-准备好提供给除我们自己以外的其他数据科学家，(Iii)需要重大改进 对于PUL/CGC数据表示、文档编制和前处理，因为当前的数据结构仅为 专为CAZymes和Puls领域专家设计；(4)有必要更新现有软件工具 以更便于计算机阅读的格式输出CGCS，以实现AI/ML就绪。 因此，这个AI/ML就绪项目的主要目标是开发一个一致的、标准化的和 PULS和CGCS的系统格式，使它们不仅为父R01项目，而且也为 其他数据科学家和营养学家。为了实现这一目标，我们召集了一个多学科的 研究团队包括三名教职员工、一名博士后和三名研究生。这些成员拥有所有 在营养科学和CAZymes、统计ML模型开发和生物信息学方面具有必要的专业知识 和ML应用程序开发。计划制定两个目标，包括四个子任务和四个里程碑，以解决 上述挑战，并使PUL和CGC数据格式化并以其方式记录 可供其他数据科学家和营养学家随时使用。所有支持AI/ML的数据将是免费的 在两个在线数据存储库上提供：DBCAN-PUL(http://bcb.unl.edu/dbCAN_PUL/)和DBCAN-SEQ (http://bcb.unl.edu/dbCAN_seq/).)这个项目将有助于对饮食调节的基本理解 人体微生物组与应用个性化营养研究。

PROJECT SUMMARY Carbohydrate enzyme gene clusters in human gut microbiome Our R01 parent project (R01GM140370) intends to develop four bioinformatics tools for automated annotation of CAZymes (Carbohydrate Active Enzymes) and CAZyme Gene Clusters (CGCs) in human gut microbiome. These automated tools will enhance: (i) the basic biomedical science to characterize new polysaccharide (or glycan) metabolic enzymes and polysaccharide utilization loci (PULs, gene clusters with known carbohydrate substrates) in the human gut microbiome, and (ii) the emerging personalized nutrition practice (e.g., using gut microbiome sequencing to infer if a person is a responder to certain dietary glycans or prebiotics). Two types of data are needed for AI/ML applications: (1) PULs (experimentally characterized gene clusters with known carbohydrate substrates) curated from literature, and (2) CGCs (without known carbohydrate substrates) predicted from human microbiome. Although the parent R01 project focuses on the development of new ML tools, challenges exist and additional support is needed to enable AI/ML-readiness for the data used/produced in the parent project. These challenges include: (i) the training data size is small, and more PULs await to be curated from literature which is not supported by the parent R01 project, (ii) the parent R01 project does not consider making the PULs and CGCs AI/ML-ready to other data scientists than ourselves, (iii) a significant improvement is needed for the PUL/CGC data representation, documentation, and pre-processing, as the current data structure is only designed for domain experts of CAZymes and PULs, (iv) an update of existing software tools will be necessary to output CGCs in a more computer-readable format, in order to enable AI/ML-readiness. Therefore, the major goal of this AI/ML-readiness project is to develop a consistent, standardized, and systematic format of PULs and CGCs to make them AI/ML ready to not only the parent R01 project but also to other data scientists and nutrition scientists. To achieve this goal, we have assembled a multi-disciplinary research team including three faculty, one postdoc, and three graduate students. These members have all necessary expertise in nutritional science and CAZymes, statistical ML model development, and bioinformatics and ML application development. Two Aims with four subtasks and four milestones are planned to address the aforementioned challenges and make the PUL and CGC data formatted and documented in a way that they can be readily available to other data scientists and nutrition scientists. All AI/ML-ready data will be freely available on two online data repositories: dbCAN-PUL (http://bcb.unl.edu/dbCAN_PUL/) and dbCAN-seq (http://bcb.unl.edu/dbCAN_seq/). This project will contribute to the basic understanding of dietary modulation of human microbiome and applied personalized nutrition research.