Integrative Analysis of Haplotype-Resolved Human Epigenome Maps

描述（由申请人提供）：NIH 路线图表观基因组联盟已为 100 多种细胞类型和组织生成了全面的表观基因组图谱。虽然这些图谱为表观遗传过程提供了许多新颖的见解，并帮助注释了人类基因组中的顺式调控元件，但对这些表观基因组图谱的深入分析却受到以下事实的困扰：每个表观基因组数据集实际上包含两个单倍体表观基因组的混合物，并且两者之间存在实质性差异。为了解决这个问题，我们建议将单倍型信息纳入不同人类细胞类型或组织的表观基因组分析中。我们开发了一种通过结合邻近连接和超高通量测序来重建染色体跨度单倍型的新策略，并且已经能够重建与来自超过 25 种细胞类型或组织的 368 个 Roadmap 表观基因组数据集相对应的单倍体基因组。我们将进行综合分析以实现三个具体目标：首先，我们将为来自七个个体的 25 种细胞类型和组织生成单倍型解析的表观基因组，并识别在这些组织或细胞类型中表现出转录、染色质修饰或 DNA 甲基化等位基因偏差的基因组区域。其次，我们将研究人 H1 ES 细胞和四种 ES 细胞衍生的早期胚胎谱系中的等位基因表达和远程控制机制，以确定增强子或启动子的等位基因特异性状态是否与人 ES 细胞分化过程中靶基因的等位基因转录相对应。第三，我们将研究不同人类组织和细胞类型中的等位基因表达和表观遗传修饰，并鉴定有助于等位基因表达的序列变异。我们预计这些分析将大大增进我们对人类细胞表观基因组景观和基因调控机制的了解。

DESCRIPTION (provided by applicant): The NIH Roadmap Epigenome Consortium has generated comprehensive epigenome profiles for over 100 cell types and tissues. While these maps have provided many novel insights into the epigenetic processes and helped annotate the cis-regulatory elements in the human genome, in-depth analysis of these epigenome maps is confounded by the fact that each epigenome dataset actually contains a mixture of two haploid epigenomes, and there are substantial differences between the two. To address this problem we propose to incorporate haplotype information into the analysis of epigenomes of diverse human cell-types or tissues. We have developed a new strategy to reconstruct chromosome-span haplotypes by combining proximity-ligation and ultrahigh throughput sequencing, and have been able to reconstruct haploid genomes corresponding to 368 Roadmap epigenome datasets from over 25 cell-types or tissues. We will perform integrative analysis to achieve three specific aims: First, we will generate haplotype-resolved epigenomes for 25 cell-types and tissues from seven individuals, and identify genomic regions that demonstrate allelic bias in transcription, chromatin modification or DNA methylation in these tissues or cell types. Second, we will investigate allelic gene expression and long-range control mechanisms in human H1 ES cells and four ES-cell-derived early embryonic lineages, to determine whether allele-specific states of enhancers or promoters correspond to allelic transcription of target genes during differentiation of human ES cells. Third, we will investigate allelic gene expression and epigenetic modifications in diverse human tissues and cell types, and identify sequence variants that contribute to allelic gene expression. We expect these analyses will substantially advance our knowledge of epigenomic landscape and gene regulatory mechanisms in human cells.