Structure function relationships from deep mutational scanning in human cardiomyopathy

项目总结 人类遗传变异的自然实验可以用来推断键的结构-功能关系 疾病基因。我们之前已经证明，种群规模的遗传变异数据可以 用于阐明孟德尔病致病基因的结构-功能关系 肥厚型心肌病。然而，由于个体致病变异的稀有，群体遗传学是 最终局限于理解任何一个编码区的功能重要性的目标 特定的基因。目前迫切需要试验性的替代方案。在这里，我们建议引入有针对性的 人类诱导多能干细胞来源的心肌细胞(IPSC-CM)规模的遗传变异(目标1)。 我们提出了两种互补的策略，用于对导致最常见的基因进行深度突变扫描 肥厚型心肌病、MYH7、MYBPC3和TNNT2。第一种是CRISPR-X，是胞苷的融合 脱氨酶(AID)与核酸酶失活的Cas9(DCas9)结合，并提供原位靶向突变覆盖。这个 第二，POPcode，使用尿苷酸化的基因模板和一组突变的寡核苷酸来创建等位基因文库，该文库 然后通过双整合酶盒交换(DICE)整合到基因组中。为了描述这些 细胞，我们进一步开发了一种定制的基于微流体的快速光学方法来真实地对单个细胞进行表型鉴定 时间(目标2)。根据细胞大小和一个荧光标记物预测致病性 肥大表达程序将使用我们的空间扫描方法映射到3D蛋白质结构 并对照金标准判定的患者变异数据进行测试。最后，我们将研究特定于变体的 用单细胞RNA测序评估每个变异对等位基因的影响 化学计量学和转录编程，以及用于评估肌节蛋白的蛋白质生物化学 互动和发电(目标3)。总之，我们计划对所有潜力进行全面评估 孟德尔人最常见心血管疾病最常见致病基因的编码变异 疾病。使用创新的表型工具和新的统计方法来整合人口 和细胞数据，我们的目标是了解这些基因在健康和疾病中的结构和功能， 为临床环境下遗传变异的分类提供了实验依据。

PROJECT SUMMARY The natural experiment of human genetic variation can be used to infer structure-function relationships for key disease genes. We have previously demonstrated that population-scale genetic variation data can be harnessed to illuminate structure-function relationships for genes causative of the Mendelian disease hypertrophic cardiomyopathy. However, due to the rarity of individual causative variants, population genetics is ultimately limiting to the goal of understanding the functional importance of the entire coding region of any specific gene. There is an urgent need for experimental alternatives. Here, we propose to introduce targeted genetic variation into human induced pluripotent stem cell derived cardiomyocytes (iPSC-CM) at scale (Aim 1). We propose two complementary strategies for deep mutational scanning of the most common genes causing hypertrophic cardiomyopathy, MYH7, MYBPC3 and TNNT2. The first, CRISPR-X, is a fusion of a cytidine deaminase (AID) with nuclease-inactive Cas9 (dCas9), and provides targeted mutational coverage in situ. The second, POPcode, uses a uracilated gene template and a set of mutant oligos to create an allelic library, which is then integrated into the genome using a Dual-Integrase Cassette Exchange (DICE). To characterize these cells, we further develop a custom microfluidics-based, fast optical method to phenotype single cells in real time (Aim 2). Predictions of pathogenicity according to both cell size and a fluorescence marker of the hypertrophy expression program will be mapped to 3D protein structures using our spatial scanning approach and tested against gold standard adjudicated patient variant data. Finally, we will investigate variant-specific mechanisms of disease using single cell RNA sequencing to assess the effect of each variant on allelic stoichiometry and transcriptional programming, as well as protein biochemistry to assess sarcomere protein interaction and power generation (Aim 3). In summary, we plan comprehensive evaluation of all potential coding variation in the most frequently causative genes for the most common Mendelian cardiovascular disease. Using innovative phenotyping tools and novel statistical approaches to the integration of population and cellular data, we aim to understand the structure and function of these genes in health and disease, providing an experimental basis for the classification of genetic variants in the clinical setting.