The Center for Actionable Variant Analysis; measuring variant function at scale

项目摘要 几乎所有与生命相容的约90亿种可能的单核苷酸变异都存在于78亿种中 今天活着的人。了解这些变异的影响，特别是在疾病相关蛋白中 编码基因，是理解人类生物学和使用基因组序列信息来指导 疾病的诊断和治疗。不幸的是，基因检测发现的大多数新变异都是 不确定意义的变体，意味着不存在足够的信息来明确解释变体， 致病的或良性的。意义不确定的变量不能用于指导患者护理和反映 我们对变异效应的不完全理解。为了克服这一挑战，我们开发了饱和基因组， 通过大规模平行测序（VAMP-seq）的基因编辑（SGE）和变体丰度， 变量效应，可以制造和测量大量变量的功能效应。在SGE，单身 核苷酸变异被直接编辑到基因组中，揭示了这些变异对细胞存活的影响 由于对剪接或蛋白质功能的影响，从而能够准确鉴定致病性和 良性变异VAMP-seq测量错义变体对细胞内蛋白质丰度的影响， 可以识别高达80%的致病变异。总之，SGE和VAMP-seq可以应用于至少40%的 基因产生高质量的，临床上有用的功能数据，在单核苷酸分辨率。已经，变体 由这些方法中的每一种产生的功能数据正被临床医生用来解释遗传变异。 我们提议的可行变异分析中心（CAVA）将利用SGE和VAMP-seq， 为约32个最具临床意义的基因组中约200，000个变异体提供单核苷酸变异体功能数据。 IGVF变体/元件/表型目录中的有影响力的蛋白质编码基因。为了实现这一 我们提出了四个转型目标。在目标1中，我们将选择靶基因并使用 最大化临床需求、临床影响和实用性的框架。每个靶标/检测试剂对将 在进入生产之前进行严格验证。我们将在第一年向联合会捐款， 通过制定标准、共享试剂和启动合作项目，在目标2中， VAMP-seq将在约32个基因上进行，以使用明确定义的指标跟踪高质量标准。这些 包括测定动态范围和再现性、个体测量误差以及与现有 功能数据和金标准临床数据。与我们的LIMS集成的数据分析管道将确保 可重复性，并能够仔细跟踪进度。在目标3中，我们将共享多路复用变体泛函 数据严格定义的数据共享标准和元数据将确保可重复性、可计算性和 耐久性我们将与财团合作，以达成共识，我们将相应地修改我们的计划。在 目标4，我们将使实验室能够快速建立SGE或VAMP-seq。我们将建立一个预测-评估-修改的循环 利用我们将生成的数据，并协同工作生成数据，以改进建模工作。

PROJECT SUMMARY Nearly all of the ~9 billion possible single nucleotide variants compatible with life exist among the 7.8 billion individuals alive today. Understanding the effects of these variants, especially in disease-associated protein coding genes, is central to understanding human biology and to using genome sequence information to guide the diagnosis and treatment of disease. Unfortunately, most new variants revealed by genetic testing are variants of uncertain significance, meaning insufficient information exists to definitively interpret the variant as either pathogenic or benign. Variants of uncertain significance cannot be used to guide patient care and reflect our incomplete understanding of variant effects. To overcome this challenge, we developed saturation genome editing (SGE) and variant abundance by massively parallel sequencing (VAMP-seq), multiplexed assays of variant effect that can make and measure the functional effect of massive numbers of variants. In SGE, single nucleotide variants are edited directly into the genome, revealing the effect of these variants on cell survival due to effects on splicing or protein function, thereby enabling accurate identification of both pathogenic and benign variants. VAMP-seq measures the effects of missense variants on protein abundance inside cells, and can identify up to 80% of pathogenic variants. Together, SGE and VAMP-seq can be applied to at least 40% of genes to produce high quality, clinically useful functional data at single nucleotide resolution. Already, variant functional data produced by each of these methods are being used by clinicians to interpret genetic variants. Our proposed Center for Actionable Variant Analysis (CAVA) will harness SGE and VAMP-seq to contribute single nucleotide variant functional data for ~200,000 variants in ~32 of the most clinically impactful protein coding genes to the IGVF Variant/Element/Phenotype Catalog. To accomplish this transformative goal we propose four Aims. In Aim 1, we will choose target genes and assays using a framework that maximizes clinical need, clinical impact and practicality. Each target/assay pair will be rigorously validated prior to entering production. We will contribute to the Consortium during the first year and beyond by developing standards, sharing reagents and initiating collaborative projects. In Aim 2, SGE and VAMP-seq will be performed on ~32 genes to high quality standards tracked using well-defined metrics. These include assay dynamic range and reproducibility, individual measurement error, and concordance with existing functional data and gold standard clinical data. A data analysis pipeline, integrated with our LIMS, will ensure reproducibility and enable careful progress tracking. In Aim 3, we will share the multiplexed variant functional data. Rigorously defined data sharing standards and metadata will ensure discoverability, computability and durability. We will work with the Consortium to achieve consensus and we will revise our plans accordingly. In Aim 4, we will enable labs to quickly stand up SGE or VAMP-seq. We will create a predict-evaluate-revise cycle that leverages the data we will generate and work collaboratively to generate data to improve modeling efforts.