Integrating T cell receptor features with gene expression profiles to define T cell specificity and differentiation

摘要 这项建议的目的是确定T细胞受体(TCR)序列和 人类T细胞全景的转录图谱。这项工作是由我们最近开发的 COMA算法是一种图论方法，它集成了TCR和基因表达(GEX)数据集， 以及技术进步，使得以高吞吐量并行分析这两个特征成为可能。 为回应特别关注通知NOT-AI-21-011(“现有数据集的二次分析”)提交 为了推进免疫调节和传染病研究“)，我们的提案汇集了一个团队 有成功合作记录的计算生物学家和免疫学家。我们的目标是申请 在不同的T细胞数据集上定义具有里程碑意义的TCR特征及其相关表型 人类T细胞。在第一个目标中，我们将识别、获取、预处理和标准化所有大型的、公开的 具有连锁基因表达和配对TCR序列信息的可用单细胞数据集。我们 然后对这些单独的数据集运行COGA管道，将结果与可用的研究相关联 元数据，并使这些结果可供下载。在第二个目标中，我们将执行Meta分析 在整个数据集中T细胞受体序列和T细胞转录图谱之间的关系 (1,000+供者和1,000,000+单个T细胞)。这里提议的工作的完成将为 为全面绘制人类T细胞图谱奠定了基础，并为进一步研究提供了有价值的数据集 分析工具和方法的发展。由COMA WILE确定的T细胞特征和亚群 提供对单个数据集的新见解，同时说明Gex/TCR的全球格局 协变。

ABSTRACT The objective of this proposal is to identify linkages between T cell receptor (TCR) sequences and transcriptional profiles across the human T cell landscape. This work is enabled by our recent development of the CoNGA algorithm, a graph theoretic approach that integrates TCR and gene expression (GEX) datasets, and by technological advances that have made it possible to profile both features in parallel at high throughput. Submitted in response to Notice of Special Interest NOT-AI-21-011 ("Secondary Analysis of Existing Datasets for Advancing Immune-mediated and Infectious Disease Research"), our proposal brings together a team of computational biologists and immunologists with a track record of successful collaboration. Our goal is to apply CoNGA on diverse T cell datasets to define the landmark TCR features and their correlated phenotypes in human T cells. In the first Aim, we will identify, acquire, pre-process, and standardize all large, publicly available single-cell datasets that feature linked gene expression and paired TCR sequence information. We will then run the CoNGA pipeline on these individual datasets, correlate the results with available study metadata, and make these results available for download. In the second Aim, we will perform a meta-analysis of the relationship between T cell receptor sequence and T cell transcriptional profile across the entire dataset (1,000+ donors and 1,000,000+ individual T cells). Completion of the work proposed here will lay the groundwork for a comprehensive atlas of the human T cell landscape and provide a valuable dataset for further development of analytical tools and methods. T cell features and sub-populations identified by CoNGA will provide new insight into the individual datasets while also illuminating the global landscape of GEX/TCR covariation.