Integrative computational models for functional epigenomics and transcriptional regulation

项目总结/摘要 基因表达的转录调控在许多细胞过程中起着关键作用。表观基因组学 是指研究蛋白质分子和生化因子的整体模式和动态变化， 与基因组DNA相互作用以影响染色质结构并调节基因表达。表观基因组学 弥合遗传变异和细胞表型之间的机械差距。功能识别 表观基因组学和转录调控关系是理解基础基因的关键 监管机制。高通量基因组方法已经越来越多地应用于该领域， 已经产生了大量的多水平基因组学数据来表征不同基因的分子谱， 不同系统中的细胞类型。这种基因组学研究的一个主要挑战是基于无偏模型的 计算分析和整合这些来自不同平台的高维多组学数据， 检索功能性见解。 我实验室的研究项目重点是开发定量模型和计算方法， 功能多组学数据分析。我们已经开发了几个计算模型和生物信息学 用于ChIP-seq数据分析的方法和用于功能性转录调控的预测模型， 整合公开可用的多组学数据。我们的长期愿景是，通过使用新的计算 方法学，采用统计学、物理学、数学和计算机等跨学科方法 科学，我们将能够了解人类细胞中基因调控的基本机制， 在许多疾病中的作用。具体而言，未来五年，我的实验室将主要围绕以下目标开展工作： (1)开发功能性转录调控关系和网络的准确预测模型， 多组学数据的智能集成。(2)开发用于无偏量化和分析的统计模型 染色质可及性测序（ATAC-seq和DNase-seq）数据。(3)发展计算 用于整合跨尺度整体和单细胞多组学数据以研究功能的联合分析方法 在单细胞水平上的调节动力学。与此同时，我们与一些实验室合作， 应用我们开发的计算方法研究功能表观基因组学和转录调控 在多种哺乳动物细胞系统中。我们致力于使我们开发的所有方法和算法成为 开源生物信息学软件工具、API和基于网络的资源，这些资源可供访问， 生物医学研究界。

PROJECT SUMMARY/ABSTRACT Transcriptional regulation of gene expression plays a critical role in numerous cellular processes. Epigenomics refers to the study of global patterns and dynamic changes of protein molecules and biochemical factors that interact with genomic DNA to affect the chromatin architecture and to regulate gene expression. Epigenomics bridges the mechanistic gaps between genetic variations and cellular phenotypes. Identification of functional epigenomics and transcriptional regulatory relations is essential for understanding fundamental gene regulatory mechanisms. High-throughput genomic approaches have been increasingly applied in the field and a large amount of multi-level genomics data have been generated to characterize molecular profiles of different cell types in various systems. One major challenge in such genomics studies is unbiased model-based computational analysis and integration of these high-dimensional multi-omics data from different platforms to retrieve functional insights. The research program of my lab focuses on developing quantitative models and computational methods for functional multi-omics data analysis. We have developed several computational models and bioinformatics methods for ChIP-seq data analysis and predictive models for functional transcriptional regulation by integrating publicly available multi-omics data. Our long-term vision is that by using novel computational methodologies with adapted cross-disciplinary approaches from statistics, physics, mathematics and computer science, we will be able to understand fundamental mechanisms of gene regulation in human cells and their role in many diseases. Specifically, in the next five years, my lab will mainly focus on the following objectives: (1) Developing accurate predictive models for functional transcriptional regulatory relations and networks with smart integration of multi-omics data. (2) Developing statistical models for unbiased quantification and analysis of chromatin accessibility sequencing (ATAC-seq and DNase-seq) data. (3) Developing computational methods for joint analysis for integrating cross-scale bulk and single-cell multi-omics data to study functional regulatory dynamics in a single-cell level. In the meantime, we collaborate with a few experimental labs and apply our developed computational methods for studying functional epigenomics and transcriptional regulation in a variety of mammalian cell systems. We commit to make all methods and algorithms that we develop into open-source bioinformatics software tools, APIs, and web-based resources that are accessible and useful to the biomedical research community.