Statistical Analysis Methods and Software for ChIP-seq Data

描述（由申请人提供）：高通量下一代测序（NGS）技术的出现通过允许对数十亿个碱基进行快速和廉价的测序，彻底改变了遗传学和基因组学领域。在NGS的应用中，ChIP-seq（染色质免疫沉淀后的NGS）可能是迄今为止最成功的。ChIP-seq技术使研究人员能够研究转录因子的全基因组结合和表观基因组标记的定位。这两者在细胞特异性的基因表达编程中都起着至关重要的作用；因此，它们的全基因组图谱可以显著提高我们理解和诊断人类疾病的能力。虽然ChIP-seq数据的基本分析工具正在迅速增加，但所有可用的方法都有以下一个或多个缺点。首先，他们专注于一次分析一个ChIP- seq样本。随着ChIP-seq被广泛用于表观基因组定位以了解表型变异，对能够有效处理多个样本的方法的需求正在迅速增加。其次，他们只使用与参考基因组上独特位置一致的序列读取。这阻碍了ChIP-seq对基因组高度重复区域的研究。第三，常用的ChIP-seq实验设计为每个ChIP-seq样本使用一个匹配的对照样本。这限制了对照实验的基因组覆盖范围，并影响了芯片样品中富集的检测。这也大大增加了大规模ChIP-seq研究的测序成本。该项目的目标是在三个具体目标中解决ChIP-seq分析的这些挑战：(1)从多样本推断的统计方法；(2)利用基因组中多个位置（multi-reads）的reads的概率模型；(3)控制实验用计算机池设计的开发与评价。这些项目将通过方法论开发、模拟、计算分析和实验验证的结合来完成。方法的开发和评估将使用来自ENCODE、modENCODE和RoadMap表观基因组学联盟的数据集以及来自合作者的新数据集。该项目产生的统计资源将以公开软件的形式传播，将为ChIP-seq实验的有效设计和分析提供必要的工具。

DESCRIPTION (provided by applicant): The advent of high throughput next generation sequencing (NGS) technologies have revolutionized the fields of genetics and genomics by allowing rapid and inexpensive sequencing of billions of bases. Among the NGS applications, ChIP-seq (chromatin immunoprecipitation followed by NGS) is perhaps the most successful to date. ChIP-seq technology enables investigators to study genome-wide binding of transcription factors and mapping of epigenomic marks. Both of these play crucial roles in programming of gene expression in a cell specific manner; therefore their genome-wide mapping can significantly advance our ability to understand and diagnose human diseases. Although basic analysis tools for ChIP-seq data are rapidly increasing, all of the available methods share one or more of the following shortcomings. First, they focus on analyzing one ChIP- seq sample at a time. As ChIP-seq is becoming commonly utilized in epigenome mapping to understand phenotypic variation, the demand for methods that can handle multiple samples efficiently is rapidly rising. Second, they only utilize sequence reads that align to unique locations on the reference genome. This hinders the study of highly repetitive regions of genomes by ChIP-seq. Third, commonly used designs for ChIP-seq experiments employ one matching control sample per each ChIP-seq sample. This limits the genome coverage of control experiments and impacts the detection of enrichment in ChIP samples. It also significantly contributes to increase in sequencing costs for large-scale ChIP-seq studies. The objective of this project is to address these challenges of ChIP-seq analysis in three specific aims: (1) Statistical methods for inference from multiple samples; (2) Probabilistic models for utilizing reads that map to multiple locations (multi-reads) in the genome; (3) Development and evaluation of in silico pooling designs for control experiments. The projects will be accomplished through a combination of methodological development, simulation, computational analysis, and experimental validation. Methods will be developed and evaluated using datasets from the ENCODE, modENCODE, and the RoadMap Epigenomics consortiums as well as novel datasets from collaborators. Statistical resources generated from the project, which will be disseminated in publicly available software, will provide essential tools for the efficient design and analysis of ChIP-seq experiments.