Cross-platform structural variant discovery with deep learning

通过深度学习跨平台结构变体发现

• 批准号: 10686879
• 负责人: Victoria Popic
• 金额: $ 57.59万
• 依托单位: BROAD INSTITUTE, INC.
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-01 至 2027-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10686879
• 关键词: Algorithms; Alzheimer' s Disease; Architecture; Autoimmune Diseases; Benchmarking; Cardiovascular Diseases; Categories; Clinical; Communities; Complex; Computer Vision Systems; Computer software; Consensus; Coupled; Data; Data Reporting; Data Set; Detection; Development; Diagnosis; Dimensions; Disease; Engineering; Evaluation; Formulation; Generations; Genetic; Genetic Diseases; Genetic Variation; Genome; Genotype; Goals; Hand; Hand functions; Hi-C; Human Genetics; Human Genome; Hybrids; Image; Learning; Link; Machine Learning; Malignant Neoplasms; Manuals; Medicine; Methodology; Methods; Mind; Modeling; Pattern; Performance; Play; Property; Research; Resolution; Role; Sampling; Science; Sequence Alignment; Signal Transduction; Source; Statistical Models; Structural Models; Structure; Techniques; Technology; Training; Variant; Work; autism spectrum disorder; blind; cancer genome; convolutional neural network; deep learning; deep learning model; deep neural network; design; engineering design; experimental study; flexibility; genome sequencing; genomic data; heuristics; improved; method development; model building; nervous system disorder; neural network; precision medicine; prototype; sequencing platform; simulation; tumor; variant detection; whole genome

Structural variants (SV) are a major driver of the genetic diversity and disease in the human genome and their discovery is imperative to advances in precision medicine and our understanding of human genetics. Due to revolutionary breakthroughs in whole-genome sequencing technologies, we now have access to genomic data at an unprecedented scale and resolution. However, despite tremendous effort and progress in SV calling methodology, general SV discovery still remains unsolved. Existing techniques use hand-engineered features and heuristics to model SV classes, relying heavily on developer expertise, which cannot scale to the vast diversity of SV types and sequencing platforms nor fully harness all the information available in raw sequencing data. As a result, these methods are usually tightly coupled to the properties of a particular sequencing technology and operate optimally only on certain SV types and sizes, rendering us blind to many other classes of SVs and their role in disease. Deep neural networks have the ability to learn complex abstractions automatically from the data and hence offer a promising avenue for general SV discovery. Deep learning has recently transformed the field of machine learning and led to remarkable advances in science and medicine. In this proposal we aim to leverage the potential of deep learning for the problem of SV detection. We lay out how to efficiently formulate SV detection as a deep learning task, and propose the development of a comprehensive framework to call and genotype SVs of different size and type, including complex and subclonal SVs, given data from a range of sequencing platforms. In particular, we demonstrate that state-of-the-art results can be obtained using our approach for short, linked, and long read datasets. In order to ensure that our models generalize across different datasets, an important goal of our proposal is also to assemble diverse and representative training data and perform extensive evaluation using publicly- available multi-platform datasets to accurately assess model performance. Our software will be built with extensibility and scalability in mind, and will be released, along with pretrained models and callsets, freely to the community.

结构变异（SV）是人类基因组及其遗传多样性和疾病的主要驱动因素 发现对于精准医学的进步和我们对人类遗传学的理解至关重要。由于 全基因组测序技术的革命性突破，我们现在可以快速获取基因组数据 前所未有的规模和分辨率。然而，尽管在 SV 调用方法方面付出了巨大的努力和进步， 一般的 SV 发现仍然没有解决。现有技术使用手工设计的特征和启发式方法 模型 SV 类，严重依赖开发人员的专业知识，无法扩展到 SV 类型的巨大多样性和 测序平台也不能充分利用原始测序数据中的所有可用信息。结果，这些 方法通常与特定测序技术的特性紧密结合并以最佳方式运行 仅针对某些 SV 类型和大小，使我们对许多其他类别的 SV 及其在疾病中的作用视而不见。深的 神经网络有能力从数据中自动学习复杂的抽象，从而提供 通用 SV 发现的有希望的途径。深度学习最近改变了机器学习领域 并导致了科学和医学的显着进步。在本提案中，我们的目标是利用深度学习的潜力 SV检测问题的学习。我们阐述了如何有效地将 SV 检测制定为深度学习 任务，并提出开发一个综合框架来调用和基因分型不同大小的 SV 类型，包括复杂的和亚克隆的 SV，给定来自一系列测序平台的数据。特别是，我们 证明使用我们的短读、链接读和长读方法可以获得最先进的结果 数据集。为了确保我们的模型能够在不同的数据集上泛化，我们提案的一个重要目标 还在于收集多样化且具有代表性的培训数据，并使用公开的方法进行广泛的评估 可用的多平台数据集来准确评估模型性能。我们的软件将构建为 考虑到可扩展性和可扩展性，并将与预训练模型和调用集一起免费发布给 社区。