Proteomic Stable Isotope Probing as a Novel Approach for Linking Prebiotics with Active Gut Microbiota

蛋白质组稳定同位素探测作为连接益生元与活性肠道微生物群的新方法

• 批准号: 10627914
• 负责人: Chongle Pan
• 金额: $ 35.07万
• 依托单位: UNIVERSITY OF OKLAHOMA
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-15 至 2026-05-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10627914
• 关键词: Acceleration; Address; Algorithms; Animals; Bioinformatics; Carbon Isotopes; Cloud Computing; Coculture Techniques; Communities; Complement; Complex; Complex Analysis; Computer Analysis; Data; Data Analyses; Databases; Detection; Diabetes Mellitus; Diet; Dietary Fiber; Disease; Environment; Equilibrium; Genotype; Goals; Health; Human; In Situ; In Vitro; Inulin; Knowledge; Label; Link; Machine Learning; Mass Spectrum Analysis; Measurement; Measures; Metabolic; Metabolism; Metagenomics; Methodology; Methods; Microbe; Modeling; Mus; Nutrition Disorders; Obesity; Organism; Outcome; Pathway interactions; Pattern; Peptides; Performance; Population; Probiotics; Protein Biosynthesis; Proteins; Proteomics; Public Health; Reproducibility; Research; Research Personnel; Resources; Ribosomal RNA; Role; Running; Shotguns; Specificity; Symbiosis; Techniques; Technology; Therapeutic; Time; Work; carbon fiber; computerized data processing; cost; data acquisition; data standards; deep learning algorithm; dysbiosis; experimental study; gut bacteria; gut microbes; gut microbiome; gut microbiota; hemicellulose; high throughput technology; human microbiota; improved; in vivo; innovation; large datasets; machine learning model; maltodextrin; metabolic abnormality assessment; metaproteomics; metatranscriptomics; microbial; microbial community; microbiome; microbiota; microorganism; novel; novel strategies; nutrition; parallel computer; prebiotics; prevent; protein biomarkers; stable isotope; tool; two-dimensional

PROJECT SUMMARY/ABSTRACT Characterization of the metabolic interactions between organisms is key to understanding the mechanisms of disease and symbioses between microbes and their animal hosts. Our long-term goal is to advance the applicability and accessibility of proteomic stable isotope probing (SIP) in ways that make it a valuable tool for microbiome researchers looking to measure in situ metabolic interactions of human microbiota. The objective of this proposal is to improve the performance and reproducibility of experimental measurements and accelerate the computational analysis of proteomic SIP experiments and to demonstrate the value of this method for studying the in vivo and in vitro metabolism of prebiotics by gut microbiota. Expected outcomes will represent a significant advance, because optimizing the use of prebiotics as therapeutics requires identification of the specific microorganisms capable of metabolizing prebiotics. By identifying proteins of specific taxa that are synthesized as a direct result of prebiotic assimilation, proteomic SIP will provide unambiguous links between prebiotic metabolism and the specific microorganisms responsible for this activity. We will accomplish this objective by pursuing three specific aims: 1) to increase the performance and reproducibility of mass spectrometry measurements by optimizing data-independent acquisition (DIA) methods for proteomic SIP; 2) to significantly accelerate the computing-intensive database search step by adapting the Sipros algorithm to use graphic processing units (GPUs) and cloud computing; and 3) to track in vivo and in vitro prebiotic assimilation patterns by microbial populations within simple consortia and complex natural communities. Our proposed work includes several innovations, such as the application of deep learning algorithms to improve the analysis mass spectrometry data, leveraging GPU-based parallel computing and cloud computing to accelerate the computational steps in the data analysis workflow, and using proteomic SIP for the first time to track prebiotic metabolism by gut microbes. The expected outcomes of the project include (a) a new DIA- based workflow for proteomic SIP that can identify significantly more labeled peptides at higher accuracy of enrichment estimation, (b) a new computational workflow that is faster to run, more scalable to large datasets, and more accessible to researchers, and (c) establish novel foundational knowledge on the specificity of prebiotic metabolism by microbes in the gut. These outcomes will establish proteomic SIP as a valuable -omics tool that will complement existing approaches to study the metabolism of gut microbiota, and specifically highlight its ability to investigate metabolism of prebiotics and probiotics as they relate to treating microbial dysbiosis and nutrition-related diseases.

项目总结/摘要 生物体之间代谢相互作用的表征是理解代谢机制的关键。 疾病和微生物与其动物宿主之间的共生关系。我们的长期目标是推动 蛋白质组稳定同位素探测（SIP）的适用性和可及性，使其成为一种有价值的工具， 微生物组研究人员希望测量人体微生物群的原位代谢相互作用。客观 该建议的目的是提高实验测量的性能和再现性， 加速蛋白质组学SIP实验的计算分析，并证明其价值 用于研究益生元通过肠道微生物群的体内和体外代谢的方法。预期成果将 代表了一个重大的进步，因为优化益生元作为治疗剂的使用需要识别 能够代谢益生元的特定微生物。通过鉴定特定分类群的蛋白质， 作为益生元同化的直接结果合成，蛋白质组学SIP将提供明确的联系 益生元代谢和负责这种活动的特定微生物之间的关系。要全面完成 通过追求三个具体目标来实现这一目标：1）提高质量的性能和再现性， 通过优化蛋白质组SIP的数据独立采集（DIA）方法进行光谱测量; 2） 通过调整Sipros算法来显著加速计算密集型数据库搜索步骤， 使用图形处理单元（GPU）和云计算;以及3）跟踪体内和体外益生元 在简单的财团和复杂的自然社区内的微生物种群的同化模式。我们 提出的工作包括几项创新，例如应用深度学习算法来改善 分析质谱数据，利用基于GPU的并行计算和云计算， 加速数据分析工作流程中的计算步骤，并首次使用蛋白质组学SIP， 追踪肠道微生物的益生元代谢该项目的预期成果包括：（a）一个新的《国家信息和通讯法》， 基于蛋白质组学SIP的工作流程，可以以更高的准确度识别更多的标记肽， 富集估计，（B）新的计算工作流程，其运行更快，更可扩展到大数据集， 研究人员更容易获得，以及（c）建立新的基础知识的特异性， 肠道中微生物的益生元代谢。这些结果将建立蛋白质组学SIP作为一个有价值的 一种工具，将补充现有的方法来研究肠道微生物群的代谢，特别是 突出了其研究益生元和益生菌代谢能力，因为它们涉及治疗微生物 生态失调和营养相关疾病。

项目成果

MetaLP: An integrative linear programming method for protein inference in metaproteomics.

• DOI: 10.1371/journal.pcbi.1010603
• 发表时间: 2022-10
• 期刊: PLOS COMPUTATIONAL BIOLOGY
• 影响因子: 4.3
• 作者: Feng, Shichao;Ji, Hong-Long;Wang, Huan;Zhang, Bailu;Sterzenbach, Ryan;Pan, Chongle;Guo, Xuan
• 通讯作者: Guo, Xuan

FineFDR: Fine-grained Taxonomy-specific False Discovery Rates Control in Metaproteomics.

FineFDR：元蛋白质组学中细粒度分类特异性错误发现率控制。

• DOI: 10.1109/bibm55620.2022.9995401
• 发表时间: 2022
• 期刊: Proceedings. IEEE International Conference on Bioinformatics and Biomedicine
• 作者: Wang,Shengze;Feng,Shichao;Pan,Chongle;Guo,Xuan
• 通讯作者: Guo,Xuan

LINA: A Linearizing Neural Network Architecture for Accurate First-Order and Second-Order Interpretations.

• DOI: 10.1109/access.2022.3163257
• 发表时间: 2022
• 期刊: IEEE ACCESS
• 影响因子: 3.9
• 作者: Badre, Adrien;Pan, Chongle
• 通讯作者: Pan, Chongle

Islet autoantibody seroconversion in type-1 diabetes is associated with metagenome-assembled genomes in infant gut microbiomes.

1 型糖尿病中的胰岛自身抗体血清转化与婴儿肠道微生物组中宏基因组组装的基因组相关。

• DOI: 10.1038/s41467-022-31227-1
• 发表时间: 2022-06-21
• 期刊: Nature communications
• 影响因子: 16.6

Explainable multi-task learning improves the parallel estimation of polygenic risk scores for many diseases through shared genetic basis.

• DOI: 10.1371/journal.pcbi.1011211
• 发表时间: 2023-07
• 期刊: PLoS computational biology
• 影响因子: 4.3