Scaling up computational genomics with tree sequences

用树序列扩展计算基因组学

基本信息

批准号：
10471496
负责人：
PETER Lochhead RALPH
金额：
$ 55.68万
依托单位：
UNIVERSITY OF OREGON
依托单位国家：
美国
项目类别：
财政年份：
2021
资助国家：
美国
起止时间：
2021-09-24 至 2023-08-31
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/10471496
关键词：
Address Affect Algorithmic Software Algorithms Architecture Area Base Sequence Collection Communities Complex Computer software Computing Methodologies Data Data Compression Data Set Development Disease Ecology Ensure Epidemiology Etiology Evolution Genealogical Tree Genealogy Generations Genetic Genetic Processes Genetic Recombination Genetic Variation Genome Genomics Genotype Goals Haplotypes Health Health Benefit Human Human Genetics Human Genome Individual Internet Libraries Maps Methods Modeling Modernization Mutation Performance Phase Phenotype Population Population Genetics Population Sizes Positioning Attribute Process Production Recording of previous events Records Research Running Sample Size Sampling Statistical Data Interpretation Structure Testing Time Training Trees Tsunami Validation Variant Work algorithm development base computer framework cost data format data reuse data structure deep learning design experience frontier genome-wide genomic data human disease improved interoperability learning strategy member multicore processor next generation novel novel strategies open source operation scale up sequence learning simulation statistics structural genomics success supervised learning whole genome

项目摘要

Project Summary/Abstract Increasing sample size is a tremendously important factor in building our understanding of the genetics of human disease. As we discover that more and more diseases have a complex web of genetic causation, we need larger and larger genetic datasets to disentangle them, and to ultimately produce successful therapies. Driven in part by this need, the community is now assembling vast collections of human genome sequences, and millions of samples will soon be commonplace. Nonhuman datasets, with applications in epidemiology, ecology, and evolution, will not be far behind. There is a profound problem, however: our computational methods for storing, processing, simulating, and analyzing genomic data are lagging far behind our ability to collect such data. The algorithms and data structures underlying today's computational methods were designed for thousands of samples, not millions, and we are in danger of being overwhelmed by the impending tsunami of data. Without a fundamental change in how we store and process genomic data, we will either not fully tap the potential of the data we collect, or the computational costs will be astronomical – or both. Our proposal addresses this critical need by focusing on a new data structure: the succinct tree sequence. This data structure (the “tree sequence”, for brevity) encodes genetic variation data using the population ge- netics processes that produced the data itself – by representing variation among contemporary samples via mutations on the branches of the underlying genealogical trees. This yields extraordinary levels of data com- pression, with ﬁle sizes hundreds of times smaller than current community standards. Since the tree sequence was introduced in 2016 it has led to performance increases of 2–4 orders of magnitude in the diverse applica- tions of genome simulation, calculation of statistics, and ancestry inference. Such sudden leaps in computa- tional performance are vanishingly rare, and only possible through deep algorithmic advances. Our research plan builds on the extraordinary successes of tree sequence methods so far, scaling up three crucial layers of computational genomics: analysis, simulation, and inference. First, we will continue our development of highly efﬁcient tree-sequence-based methods for fundamental operations in statistical and population genetics. Second, we will scale up genome simulations by integrating tree sequence methods into complex forward-time simulations, utilizing modern, multicore processors. Third, we will combine efﬁcient genome simulations with cutting-edge deep-learning methods to improve existing inference methods, both of tree sequences from genomic data, and of population parameters from novel tree-sequence encodings of genotype data. Together, we aim to revolutionize the way we work with population genetic variation data, and how we use it to understand human health and evolutionary processes. Our experienced, interdisciplinary team is committed to producing rigorously tested and validated software and accessible, interoperable, and reusable data formats through inclusive and open development.

项目概要/摘要增加样本量是建立我们对遗传学的理解的一个极其重要的因素当我们发现越来越多的疾病具有复杂的遗传因果关系网络时，我们需要越来越大的基因数据集来解开它们，并最终产生成功的治疗方法。在这种需求的推动下，社区现在正在收集大量的人类基因组序列，数以百万计的样本将很快变得普遍，并应用于流行病学，然而，生态学和进化论也不会落后太多，但有一个深刻的问题：我们的计算能力。存储、处理、模拟和分析基因组数据的方法远远落后于我们的能力设计了当今计算方法的算法和数据结构。需要数千个样本，而不是数百万个样本，我们面临着被即将到来的海啸淹没的危险如果我们存储和处理基因组数据的方式没有发生根本性的改变，我们要么无法充分利用数据。我们收集的数据的潜力，或者计算成本将是天文数字——或者两者兼而有之。我们的提案通过关注一种新的数据结构来解决这一关键需求：简洁的树序列。该数据结构（为简洁起见，称为“树序列”）使用群体基因编码遗传变异数据产生数据本身的网络过程——通过表示当代样本之间的变化潜在遗传树分支上的突变会产生非凡的数据水平。压缩，文件大小比当前社区标准小数百倍。于 2016 年推出，它使各种应用程序的性能提高了 2-4 个数量级基因组模拟、统计计算和祖先推断的突然飞跃。性能表现极其罕见，只有通过深入的算法进步才有可能实现。我们的研究计划建立在树序列方法迄今为止取得的非凡成功的基础上，扩大了三个计算基因组学的关键层：分析、模拟和推理首先，我们将继续我们的研究。开发基于树序列的高效方法，用于统计和统计中的基本操作其次，我们将通过将树序列方法整合到基因组模拟中来扩大基因组模拟的规模。第三，我们将利用现代多核处理器进行复杂的前向时间模拟。使用尖端深度学习方法进行基因组模拟，以改进现有的推理方法，来自基因组数据的树序列，以及来自新的树序列编码的群体参数我们共同致力于彻底改变我们处理群体遗传变异数据的方式，以及我们如何使用它来了解人类健康和进化过程。我们经验丰富的跨学科团队致力于生产经过严格测试和验证的软件通过包容性和开放性的开发，实现可访问、可互操作和可重用的数据格式。

项目成果

期刊论文数量（7）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

A general and efficient representation of ancestral recombination graphs.

祖先重组图的通用且有效的表示。

DOI：
发表时间：
2024-04-23
期刊：
影响因子：
0
作者：
Wong, Yan;Ignatieva, Anastasia;Koskela, Jere;Gorjanc, Gregor;Wohns, Anthony Wilder;Kelleher, Jerome
通讯作者：
Kelleher, Jerome

Looking forwards and backwards: dynamics and genealogies of locally regulated populations.

向前和向后展望：当地管制人口的动态和谱系。