Milk-Omics: Systems Biology of Human Milk and Its Links to Maternal and Infant Health

项目总结 母乳对婴儿的健康有许多好处，但造成这些好处的成分 每个婴儿遗骸独特的牛奶“配方”的益处和基本分子机制 缺乏。基因组学进展开创了探索人类其他组织的细胞和分子生物学的先河 为理解生物系统提供了强大的策略，但在牛奶研究中没有得到充分利用。是时候了 在母乳研究的新时代，专注于深入询问母体乳腺细胞基因组学，在 与母体临床和行为因素相互作用，影响乳汁成分。此外，其范围还包括 牛奶成分的正常变异还有待确定。概述牛奶中标准变异的范围 是支持早产儿营养需求的下一代母乳强化技术的关键 和其他感兴趣的临床人群。拟议项目的主要目标是产生一个 在健康母亲及其足月婴儿的背景下，系统地看待母乳。因此， 拟议的工作直接响应NIH/NICHD RFA-022-020，“作为生物系统的母乳”。这个 这项研究利用了400名母婴的丰富表型队列的现有数据、牛奶和粪便样本 基于令人信服的初步数据，确定了塑造牛奶的新基因序列变异 乳汁代谢组、脂组和微生物组变异与婴儿生长发育的关系及基因表达 和认知力。具体目标1：确定影响母乳基因的母体遗传和临床因素 表情。我们将识别牛奶转录组的新遗传决定因素，并评估其潜力 通过孕期体重增加、饮食和其他临床因素改变这些遗传关联。单细胞 RNA测序将为细胞类型组成的表征提供额外的必要信息 在人类乳汁中。具体目标2：描述规范的母乳生物系统的主要特征及其 彼此之间的互动。该项目扩展了队列中现有的牛奶组学数据，包括牛奶 微生物组、低聚糖、代谢组学和脂类组学，与基因组数据整合 在目标1下制作。将使用已建立的和新颖的机器学习技术来表征 母乳的这些关键特征之间的相互作用网络和相关结构。具体目标3： 确定牛奶生物系统如何与婴儿肠道微生物群和健康的变化有关。牛奶多组体 网络将与婴儿生长、身体组成和肠道微生物群从出生到6岁的变化保持一致 在上述队列中的400名婴儿后代中有400个月。统计和机器学习技术将 确定与正常婴儿生长和肠道微生物群一致的乳汁系统变化范围 发展。对于150名婴儿的子集，我们还将纳入创新的早期生活认知评估 用于探索性分析。最后，我们将开发一个用于可视化结果数据的在线门户，从而实现 其他研究人员调查感兴趣的特定特征和关系。

PROJECT SUMMARY Human milk confers numerous infant health benefits, yet the compositional elements responsible for these benefits and the fundamental molecular mechanisms shaping the unique milk “recipe” for each infant remains lacking. Genomic advances pioneered to probe the cellular and molecular biology of other human tissues provide powerful strategies to understand biological systems, but are underutilized in milk research.. It is time for a new era of human milk research focused on deep interrogation of maternal mammary cell genomics, in interaction with maternal clinical and behavioral factors, in shaping milk composition. Furthermore, the scope of normal variation in milk composition is yet to be established. Outlining the scope of normative variation in milk is key to next-generation human milk fortification techniques to support the nutritional needs of preterm infants and other clinical populations of interest. The primary objective of the proposed project is to generate a systems-level view of human milk in the context of healthy mothers and their term infants. As such, the proposed work is directly responsive to NIH/NICHD RFA-022-020, “Human Milk as a Biological System”. The study leverages existing data, milk, and fecal specimens from a richly phenotyped cohort of 400 mother-infant dyads, and is based on compelling preliminary data identifying novel genetic sequence variation shaping milk gene expression, and relationships of milk metabolomic, lipidomic, and microbiomic variation to infant growth and cognition. Specific Aim 1: to identify maternal genetic and clinical factors that shape human milk gene expression. We will identify novel genetic determinants of the milk transcriptome and assess potential modification of these genetic associations by gestational weight gain, diet, and other clinical factors. Single-cell RNA-sequencing will provide additional necessary information for characterization of the cell type composition within human milk. Specific Aim 2: to describe key features of the normative human milk biosystem and their interactions with one another. The project expands on existing milk omics data from the cohort, including milk microbiomes, oligosaccharides, metabolomics, and lipidomics, to be integrated with the genomic data produced under Aim 1. Established and novel machine learning techniques will be used to characterize interaction networks and correlational structures among these key features of human milk. Specific Aim 3: to establish how the milk biosystem is related to variation in infant gut microbiomes and health. Milk multi-‘omic networks will be aligned with infant growth, body composition, and gut microbiome variation from birth to 6 months in the 400 infant offspring from the above cohort. Statistical and machine learning techniques will define the scope of milk system variation consistent with normal infant growth and gut microbiome development. For a subset of 150 infants, we will also incorporate innovative early life cognitive assessments for exploratory analyses. Finally, we will develop an online portal for visualizing the resulting data, enabling other researchers to investigate specific features and relationships of interest.