Project I - Transcriptomic Analysis of Structural Birth Defects in Mouse Developmental Mutants

项目摘要 在这个项目中，我们计划应用一个强大的和可扩展的组合索引技术来表征 携带已知或可能导致以下突变的小鼠中单细胞水平的基因表达： 结构性出生缺陷我们建议评估单细胞表达是否可以用作表型; 具体地说，我们旨在组织这些数据集，以评估是否存在单细胞基因特征， 基于发育信号传导的假定途径促进对突变株系分组的表达 扰动该分析将通过项目2中提出的解剖分析进行补充。 鉴于这种方法的新奇，我们将首先分析10个株系，这些株系在Shh中的基因发生突变。 信号通路，以便将单细胞转录组学数据与充分研究的发育 表型为了最大限度地增加发现新基因的机会，我们还将研究具有以下特征的新基因： 以前没有关于人类结构性出生缺陷的注释。具体而言，通过分析 在人类外显子组测序数据中，我们已经鉴定了可能是单倍不足的基因的大群组;即， 当杂合子无效时，它们与存活不相容。我们还培育了一个杂合子 选择（shet）统计数据与人类疾病严重程度非常相关。我们的目标是描述75 来自具有有限功能注释的顶部五分位数表集的线;这些基因将被选择为a） 基于胚胎发生期间单细胞表达的证据（Cao et al. 2019），它们是新的细胞， 型特异性指标基因或B）是已知的致死基因（在小鼠中），其具有高频率的蛋白质 交互. 作为这项工作的一部分，我们将开发生物信息学工具，以促进不同数据集之间的比较。 这些可以识别具有发育信号传导的常见异常的突变株系，以及潜在的 作为一种手段来理解人类先天性异常的机械基础。

PROJECT SUMMARY In this project we plan to apply a powerful and scaleable technique of combinatorial indexing to characterize gene expression at the single-cell level in mice carrying mutations that are either known to or likely to result in structural birth defects. We propose to assess whether single-cell expression can be utilized as a phenotype; specifically, we aim to organize these data sets to assess whether there are signatures of single-cell gene expression that facilitate grouping mutant lines based on presumptive pathways of developmental signaling perturbation. This analysis will be complemented by the anatomical analysis that is proposed in Project 2. Given the novelty of this method, we will initially analyze 10 lines that are mutated for genes in the Shh signaling pathway, in order to correlate single-cell transcriptomic data with well-studied developmental phenotypes. To maximize the opportunity for new gene discovery, we will also examine novel genes that have not been previously annotated with respect to human structural birth defects. Specifically, using an analysis of human exome sequencing data, we have identified a large cohort of genes that are likely haploinsufficient; i.e., they are not compatible with survival when heterozygous null. We have furthermore developed a heterozygote selection (shet) statistic that correlates remarkably well with human disease severity. We aim to characterize 75 lines from the top quintile shet set that have limited functional annotation; these genes will be chosen either a) based on evidence from single-cell expression during embryogenesis (Cao et al. 2019) that they are novel cell- type-specific index genes or b) are known lethal genes (in mice) that have a high frequency of protein interactions. As part of this effort we will develop bioinformatic tools to facilitate comparisons across different datasets. These can identify mutant lines with common abnormalities of developmental signaling, as well as potentially serving as a means to understand the mechanistic basis for human congenital abnormalities.