High-resolution map of human germline mutation patterns and inference of mutagenic mechanisms

 描述(申请人提供)：突变是基因变异的最终来源，也是进化的驱动力之一。突变亚型之间的绝对突变率和相对突变率都随基因组波动，受相邻核苷酸基序和局部特征(如GC含量和复制时机)的影响。表征突变模式的区域变异对于理解基因组进化和识别导致遗传病的变异至关重要。然而，突变率和分子光谱很难在高分辨率、全基因组和无偏见的方式下测量。基于常见变异和物种间替换的估计受到自然选择、种群人口历史和有偏见的基因转换(BGC)的混淆。依赖单基因疾病发病率或通过TRIO测序发现从头变异的方法可以提供全球趋势信息，但不能提供足够的数据来评估精细的局部参数。这项研究将通过使用极其罕见的变异(ERV)作为新的数据源来表征人类最近生殖系变异的模式，从而克服这些限制。ERV在本研究中定义为30,000个样本中的单身个体，正在通过对人群样本的大规模全基因组测序(WGS)获得。与常见的变异或替代不同，ERV是最近出现的，在很大程度上不受选择、BGC等的影响。我们将在20-30倍的覆盖范围内分析在30,000名受试者中观察到的2-3亿个单一变异。ERV亚型的区域分布将建立人类生殖系突变的速率和谱的定量图谱，几乎不会因选择而改变。我们将与研究界共享这一资源，并将其应用于确定当地基因组特征和表观基因组属性的影响。我们将利用ERV和更高频率的变异体(多态和替换)之间的系统偏离来推断选择的局部效应，这可能会揭示基因组中迄今未知的功能区域。通过比较ERV中的突变特征与在不同癌症中观察到的体细胞变异中的突变特征，我们将把不同的突变特征归因于已知的生化过程，从而推断人类基因组中新的生殖系突变的主要贡献者。这份特定亚型的图谱还将被用来预测观察到基因组中每一个可能的单碱基突变的可能性，从而有助于解释人类疾病的候选因果变异。我们将评估欧洲裔美国人、非裔美国人和拉美裔美国人之间的突变模式差异，并通过寻找在周围基因组区域显示ERV数量增加的功能破坏性突变，潜在地识别在人类传播保真度中发挥作用的已知和先前未知的“突变子”基因，寻求发现生殖系突变率的遗传修饰物。这项研究将为研究人类生殖系突变的发生和维持提供必要的资源。了解这一基本过程将是更深入地了解人类进化和疾病的基础。

 DESCRIPTION (provided by applicant): Mutations are the ultimate source of genetic variation and one of the driving forces of evolution. Both the absolute mutation rate and the relative rate among mutation subtypes fluctuate along the genome, affected by adjacent nucleotide motifs and local features such as GC content and replication timing. Characterizing regional variation of mutation patterns is critical for understanding genome evolution and to identify variants causing genetic diseases. However, mutation rate and molecular spectrum are difficult to measure at high resolution, genomewide, and in an unbiased fashion. Estimates based on common variants and between- species substitutions are confounded by natural selection, population demographic history, and biased gene conversion (BGC). Methods relying on incidence rates of monogenic diseases or finding de novo variants by trio sequencing can inform global trends, but do not provide sufficient data to assess fine-scale local parameters. This study will overcome these limitations by using the extremely rare variants (ERVs) as a new data source to characterize patterns of recent germline variation in humans. ERVs, defined in this study as singletons in 30,000 samples, are becoming available via large-scale whole-genome sequencing (WGS) of population samples. Unlike common variants or substitutions, ERVs arose very recently and are largely unaffected by selection, BGC, etc. We will analyze 200-300 million singleton variants observed in 30,000 subjects at 20-30X coverage. The regional distribution of ERV subtypes will establish a quantitative atlas of the rate and spectrum of human germline mutations mostly unaltered by selection. We will share this resource with the research community and apply it to determine the impact of local genomic features and epigenomic attributes. We will use the systematic departures between ERVs and variants of higher frequencies (polymorphisms and substitutions) to infer local effects of selection, and this may uncover hitherto unknown functional regions of the genome. By comparing mutation signatures in ERVs with those in somatic variations observed in diverse cancers we will attribute distinct mutational signatures to known biochemical processes and thus infer the major contributors to new germline mutations in the human genome. This subtype-specific atlas will also be used to predict the probability of observing every possible single-base mutation in the genome, thus facilitating the interpretation of candidate causal variants of human diseases. We will assess mutation pattern differences among European Americans, African Americans and Latinos, and seek to discover genetic modifiers of germline mutation rate by finding functionally damaging mutations that show increased ERV counts in the surrounding genomic region, potentially identifying both known and previous unknown "mutator" genes that play a role in transmission fidelity in humans. This research will provide an essential resource to study the genesis and maintenance of germline mutations in humans. Understanding such a fundamental process will be the basis for a deeper understanding of human evolution and diseases.