Integrative analysis tools to dissect cell-type specific transcriptional programs

描述（由申请人提供）：如何建立和维持细胞类型特异性基因表达程序是分子生物学中的一个基本问题。在哺乳动物细胞中，已经编目了数百种序列特异性转录因子，并且它们以细胞类型特异性和组合占据模式结合其靶基因的调控区。此外，产生不同细胞谱系的发育程序伴随着复杂的染色质重塑。越来越多的证据表明，细胞类型特异性基因的调控区域可能经常被建立，有时在发育的早期阶段通过染色质标记“保持平衡”。然而，基因调控区的详细特征，包括其在早期祖细胞中的初始建立，其染色质状态的动态，以及由多个转录因子组合控制基因转录输出，仅研究了少数发育重要基因。 该项目的目标是开发新的综合计算方法，利用大量的下一代测序数据集，从根本上推进我们对细胞类型特异性转录程序的理解。我们将开发整合的计算分析方法，用于（1）从ChIP-seq和DNase-seq中了解转录因子结合的序列和染色质决定簇;（2）在所有可用的细胞类型中使用DNase-seq绘制人和小鼠基因组中所有调控区的染色质可及性的景观，剖析它们在早期祖细胞中的染色质状态的平衡，并提取控制它们在分化中获得和丧失的序列密码;和（3）将细胞类型特异性基因表达程序建模为染色质状态、转录因子结合和调控序列分析的函数。我们将把我们的计算方法开发与有针对性的实验验证结合起来，包括基因座特异性和全基因组分析。

DESCRIPTION (provided by applicant): How cell-type specific gene expression programs are established and maintained is a fundamental question in molecular biology. In mammalian cells, hundreds of sequence-specific transcription factors have been catalogued, and they bind the regulatory regions of their target genes in cell-type specific and combinatorial occupancy patterns. Moreover, the developmental programs that generate different cell lineages are accompanied by complex chromatin remodeling. Increasing evidence suggests that the regulatory regions of cell-type specific genes may often be established and sometimes "poised" by chromatin marks at earlier stages in development. However, the detailed characterization of gene regulatory regions-including their initial establishment in earlier progenitor cells, the dynamics of their chromatin state, and the combinatorial control of gene transcriptional output by multiple transcription factors-has only been studied for a handful of developmentally important genes. The goal of this project is to develop new integrative computational methods that exploit massive next- generation sequencing data sets to fundamentally advance our understanding of cell-type specific transcriptional programs. We will develop integrative computational analysis methods for (1) learning the sequence and chromatin determinants of transcription factor binding from ChIP-seq and DNase-seq; (2) mapping the landscape of chromatin accessibility of all regulatory regions in the human and mouse genomes using DNase-seq across all available cell types, dissecting the poising of their chromatin state in earlier progenitor cells, and extracting the sequence code governing their gain and loss in differentiation; and (3) modeling cell-type specific gene expression programs as a function of chromatin state, transcription factor binding, and regulatory sequence analysis. We will couple our computational methods development with targeted experimental validation, including both locus-specific and genome-wide assays.