Statistical Methods for Next-Generation Sequence Data

描述（由申请人提供）：该项目的广泛，长期目标涉及开发新的统计方法和计算工具，用于重要生物学问题和实验激发的大规模下一代序列（NGS）数据的统计和概率建模。当前项目的具体目标是开发用于分析NGS数据的新的统计模型和计算方法，重点是用于发现生殖系DNA中拷贝数变异（CNV）的稳健方法，开发用于识别单样本和多样本RNA-seq数据上的替代外显子使用的通用对数线性模型，允许短读段测序速率的不均匀性，基于染色质免疫沉淀和高通量测序（ChIP-seq）数据开发用于鉴定组蛋白修饰位点的新型非参数统计方法，以及用于分析来自人类微生物组研究的宏基因组数据的新型方法。这些问题都是由PI与宾夕法尼亚大学调查人员的密切合作引起的。该方法基于生物学见解和高维数据分析方法的新集成，包括稀疏结构信号的检测和识别，基于小波的非参数回归和非参数假设检验以及树结构协变量的惩罚回归分析。新方法可以应用于不同类型的NGS数据，并将理想地促进各种复杂的人类疾病和生物过程的基因和生物途径的识别。该项目还将调查这些方法的鲁棒性、能力和效率，并将其与现有方法进行比较。此外，该项目将开发实用可行的计算机程序，以实施所提出的方法，通过应用于与非洲人群中的CNV和RNA-seq分析相关的NGS数据集来评估这些方法的性能，过氧化物酶体增殖物激活物受体（PPAR）3与脂肪分化和胰岛素抵抗的联系以及饮食对人类微生物组的影响。这里提出的工作将有助于统计方法来建模超高维下一代序列数据，研究复杂的表型和生物系统，并提供对各种数据集所代表的每个生物领域的见解。根据这项资助开发的所有项目和详细的文件将免费提供给感兴趣的研究人员。

DESCRIPTION (provided by applicant): The broad, long-term objective of this project concerns the development of novel statistical methods and computational tools for statistical and probabilistic modeling of large-scale next-generation sequence (NGS) data motivated by important biological questions and experiments. The specific aim of the current project is to develop new statistical models and computational methods for analysis of NGS data, focusing on robust methods for discovering copy number variants (CNVs) in germline DNAs, development of a general log-linear model for identifying alternative exon usages on one- and multi-sample RNA-seq data allowing for non-uniformity on short- read sequencing rates, development of novel nonparametric statistical methods for identifying histone modification sites based on the chromatin immunoprecipitation and high-throughput sequencing (ChIP-seq) data, and novel methods for analysis of metagenomic data from human microbiome studies. These problems are all motivated by the PI's close collaborations with Penn investigators. The methods hinge on novel integration of biological insights and methods for high dimensional data analysis, including detection and identification of sparse structured-signals, wavelet-based nonparametric regression and nonparametric hypothesis testing and penalized regression analysis for tree-structured covariates. The new methods can be applied to different types of NGS data and will ideally facilitate the identifications of genes and biological pathways underlying various complex human diseases and biological processes. The project will also investigate the robustness, power and efficiencies of these methods and compare them with existing methods. In addition, this project will develop practical and feasible computer programs in order to implement the proposed methods, to evaluate the performance of these methods through applications to NGS data sets related to CNV and RNA-seq analysis in African populations, linkage of peroxi- some proliferator activator receptor (PPAR)3 and adipose differentiation and insulin resistance and effects of diets on human microbiome. The work proposed here will contribute statistical methodology to modeling ultra-high dimensional next-generation sequence data and to studying complex phenotypes and biological systems and offer insights into each of the biological areas represented by the various data sets. All programs developed under this grant and detailed documentation will be made available free-of-charge to interested researchers.