Comprehensive Informatic Analyses of AML Genomes and Epigenomes

摘要 在过去的十年里，基因组研究揭示了急性髓系白血病中出现的突变的图景 白血病(AML)。在每一种情况下，这些突变中的一个代表了甚至是该肿瘤的起源。正在启动 当突变发生在造血干细胞/祖细胞(HSPC)中时，它们可以创造一种健康优势， 导致克隆性造血，这种状态增加了白血病转化的可能性。虽然我们 已经表征了这些启动突变的一些后果，包括转录变化和 DNMT3A突变的局灶性低甲基化表型，我们对这些变化的理解 推广AML在很大程度上是不完整的。由于这些启动事件发生在肿瘤的每个细胞中，不同于后来的- 获得性亚克隆事件，它们也是有吸引力的治疗靶点。因此，这项研究的第一个目标 程序是定义启动突变导致AML的分子机制，并使用 这些信息有助于开发新的、分子靶向的疗法。我的角色将是提取大的基因组 和表观基因组数据集到直观、可理解的模型中，生成关于 细胞转化为AML的过程。要做到这一点，将需要创造性的算法开发和 统计建模，我精通的生物信息学领域。然后，这种知识可以引导我们 确定新疗法的目标和方法的优先顺序。 我们研究计划的第二个主题是找出急性淋巴细胞白血病“缺失”的原因。 突变。大多数启动突变本身不足以产生明显的AML，但约5%的AML 病例似乎没有明显的合作突变。因此，我们将使用新技术和 从基因组的“暗物质”中寻找与AML相关的基因组和 表观基因组变化，使用长读测序来查询以前无法解析的基因组部分， 全基因组测序以探索非编码区和结构变异，以及算法开发 揭示急性髓系白血病中难以确定的事件。 由于我的跨学科背景，我在设计算法和统计模型方面拥有专业知识，如 以及对急性髓系白血病生物学的深刻理解。我的培训，经验和记录，生产和 有影响力的研究使我非常适合推动这项研究的信息学和分析方面 程序前进。

Abstract Over the last decade, genomic studies have revealed the landscape of mutations that appear in Acute Myeloid Leukemias (AML). In each case, one of these mutations represents the initiating even for that tumor. Initiating mutations can create a fitness advantage when they occur in hematopoietic stem/progenitor cells (HSPC), resulting in clonal hematopoeisis, a state that increases the likelihood of leukemic transformation. Although we have characterized some consequences of these initiating mutations, including transcriptional changes and focal hypomethylation phenotypes for the DNMT3A mutations, our understanding of how these changes promote AML is largely incomplete. Since these initiating events occur in every cell of the tumor, unlike later- acquired subclonal events, they also are attractive targets for therapy. Thus, the first goal of this research program is to define the molecular mechanisms by which initiating mutations cause AML, and to use this information to develop novel, molecularly-targeted therapies. My role will be to distill large genomic and epigenomic datasets into intuitive, comprehensible models, generating testable hypotheses about the process of cellular transformation into AML. Doing so will require creative algorithmic development and statistical modeling, areas of bioinformatics in which I am proficient. This knowledge can then guide us in prioritizing targets and approaches for novel therapeutics. A second theme of our research program is to identify the reasons for progression of AMLs with "missing" mutations. Most initiating mutations are insufficient to produce overt AML on their own, but about 5% of AML cases appear to have no clear cooperating mutations. We will therefore use new technologies and analytical approaches to search the "dark matter" of the genome for AML-relevant genomic and epigenomic changes, using long-read sequencing to query previously unresolvable portions of the genome, whole genome sequencing to explore non-coding regions and structural variation, and algorithmic development to reveal difficult to ascertain events in AML. Because of my interdisciplinary background, I have expertise in designing algorithms and statistical models, as well as a deep understanding of the biology of AML. My training, experience, and record of productive and impactful research make me uniquely suited to push the informatics and analysis aspects of this research program forward.