Statistical Methods for Analyzing Birth Defects Cohorts

项目摘要 出生缺陷给全球家庭和社会造成重大的健康和经济负担。近年来， 生物技术的进步，如下一代测序，有助于识别许多疾病， 导致出生缺陷和儿童癌症的基因。虽然已确定的基因只能解释一小部分 这些进展表明，有希望确定更多的出生缺陷原因， 通过强大的统计方法从测序数据分析中获得基因。NIH共同基金 建立了加布里埃拉米勒儿童第一儿科研究计划（儿童第一），以“发展一个儿科 由基因组序列和表型数据填充的研究数据资源，将对 研究儿童癌症和/或结构性出生缺陷遗传学的研究人员社区。”的 该项目的最终目标是开发、实施和应用新型统计方法来提高能力 利用儿童第一数据中心的数据， 资源中心，并使开发的工具提供给科学界。这将是 实现三个具体目标。首先，我们将制定一个统计框架， 考虑不同的疾病模型-包括从头突变和罕见的遗传变异-以更多 从全外显子组测序数据中有效鉴定致病基因。第二，我们将发展 统计方法来量化共享的新生突变对不同出生缺陷的贡献程度 以及可以利用这种共享遗传学来识别致病基因的方法。三是经过 评估我们开发的方法的性能，我们将实施这些方法，并将其应用于 出生缺陷队列目前可在儿童第一数据资源中心以及其他数据集， 将在未来添加。我们还将向科学界传播该软件。在完成 我们的目标是，我们将贡献新的统计工具来分析出生缺陷队列， 发现不同出生缺陷的基因和途径。

Project Summary Birth defects cause significant health and economic burdens to families and societies globally. In recent years, advances in biotechnologies, such as next-generation sequencing, have helped to identify many disease- causing genes for birth defects and childhood cancers. Although the identified genes only explain a small proportion of the cases, these advancements demonstrate the promise of identifying more birth defect-causing genes from the analysis of sequencing data through powerful statistical methods. The NIH Common Fund established the Gabriella Miller Kids First Pediatric Research Program (Kids First) to “develop a pediatric research data resource populated by genome sequence and phenotype data that will be of high value for the communities of investigators who study the genetics of childhood cancers and/or structural birth defects.” The ultimate goal of this project is to develop, implement, and apply novel statistical methods to improve the power of identifying genes causing birth defects across a number of conditions using data from the Kids First Data Resource Center and to make the developed tools available to the scientific community. This will be accomplished through three specific aims. First, we will develop a statistical framework that can simultaneously consider different disease models – including both de novo mutations and rare inherited variants – to more effectively identify disease-causing genes from whole exome sequencing data. Second, we will develop statistical methods to quantify the degree of shared de novo mutation contributions to different birth defects and also methods that can leverage this shared genetics to identify disease-causing genes. Third, after evaluating the performance of our developed methods, we will implement these methods and apply them to the birth defect cohorts currently available at the Kids First Data Resource Center as well as other data sets that will be added in the future. We will also disseminate the software to the scientific community. In accomplishing our aims, we will contribute new statistical tools to analyze birth defects cohorts as well as make new biological discoveries of genes and pathways for different birth defects.