Long read based sequencing software for the comprehensive analysis of clinical samples

基于长读长的测序软件，用于临床样本的综合分析

• 批准号: 10009727
• 负责人: TIMOTHY J DURFEE
• 金额: $ 75万
• 依托单位: DNASTAR, INC.
• 依托单位国家: 美国
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-04-01 至 2022-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10009727
• 关键词: Biological Sciences; Candidate Disease Gene; Catalogs; Clinical; Collaborations; Computer software; Computers; DNA; DNA Resequencing; Data; Data Set; Detection; Development; Disease; Ensure; Environment; Evaluation; Feedback; Gene Family; Genes; Genome; Genomic DNA; Goals; Haplotypes; Heterozygote; Hour; Individual; Kidney Diseases; Laboratories; Length; Methods; Nature; Phase; Polishes; Population; Privacy; Process; Provider; Pseudogenes; Research Personnel; Running; Sampling; Series; Statistical Methods; Stream; Structure; Targeted Resequencing; Technology; Time; Variant; Visualization; analytical tool; annotation system; base; clinical application; clinical sequencing; cohort; contig; cost; cost effective; disorder prevention; genetic variant; genome browser; genome sequencing; genome-wide; improved; insertion/deletion mutation; instrument; interest; nanopore; next generation sequencing; novel; parallel processing; prototype; reference genome; software development; tool; trait; whole genome

The high cost and complexity of the analysis of whole genome resequencing remain prohibitive for most clinical applications. Targeted resequencing allows regions of interest to be enriched from a genomic DNA sample and sequenced to high depth allowing cost-effective identification of important variants. In combination with next- generation sequencing (NGS), the approach has been exploited to tremendous effect in identifying candidate genes and variants for an array of diseases and traits from cohorts and populations as well as individual clinical samples. However, the short read nature of NGS technologies severely limits its potential to characterize, for example, compound heterozygotes due to the lack of long range connectivity needed for haplotype phasing and structural variants (SV). Those limitations can be overcome with long read data from Pacific Biosciences (PacBio) or Oxford Nanopore Technologies (ONT). Moreover, new targeting methods tailored toward long read sequencing are being developed such that a comprehensive analysis of key regions in an individual’s genome will soon be within reach. However, an integrated software solution that is easy enough for clinical researchers to efficiently use is sorely lacking. The overall goal of this Direct to Phase II proposal is to develop commercial-grade software that produces a comprehensive catalog of annotated haplotype phased variants from clinical sequencing data and presents them to clinical researchers through a single easy-to-use application with both analytical and genome browsing capabilities, GenVision Ultra. The proposal focuses on augmenting our highly extensible XNG assembly pipeline with tools necessary for fully automated detection and annotation of all classes of variants from haplotype phased sequences. Novel adaptions to core XNG components will partition reads matching the reference from those likely representing a SV for parallel processing (Aim 1). Matching reads will be aligned to the reference using XNG while the putative SV-containing reads will be de novo assembled and annotated using our long read assembler (LRA). Reference-based alignments will be phased using a novel Bayesian classifier to produce two haplotype sequences prior to SNV/small indel calling and annotation (Aim 2). Short read polishing of the entire assembly will be available on demand. Complete small variant and SV profiles as well as the underlying assembly data will be accessible to the end user in GenVision Ultra. In addition, the application will have discrete filtering and statistical tools with which to identify genes and/or variants of interest in an individual sample or across a cohort/population (Aim 3). To ensure that the software meets the clinical sequencing market needs, Arkana Laboratories has agreed to provide ONT and Illumina sequence data from highly curated HapMap control samples processed with their kidney disease gene panels. Those real-world data sets together with expert interpretation and feedback by Arkana researchers provide an ideal environment in which to develop an outstanding software solution for this critical market (Aim 4).

全基因组重测序分析的高成本和复杂性对大多数临床来说仍然是令人望而却步的 申请。靶向重测序允许从基因组DNA样本中丰富感兴趣的区域 测序到很高的深度，从而能够经济高效地识别重要的变异。与NEXT相结合- 生成排序(NGS)，该方法已经在确定候选对象方面取得了巨大的效果 来自队列和人群以及个人临床的一系列疾病和特征的基因和变体 样本。然而，NGS技术的短读特性严重限制了其表征、 例如，由于缺乏单倍型阶段所需的远距离连接而导致的复合杂合子 结构变种(SV)。这些限制可以通过太平洋生物科学(PacBio)的长时间读取数据来克服 或牛津纳米孔技术公司(Oxford Nanopore Technologies)。此外，针对长期阅读量身定做的新目标方法 正在开发的测序技术使得对个体基因组中关键区域的全面分析 很快就会触手可及。然而，对于临床研究人员来说，一个足够容易的集成软件解决方案 要想有效地利用这些资源是非常缺乏的。 这个直接到第二阶段提案的总体目标是开发商业级软件， 来自临床测序数据的带注释的单倍型阶段性变异的综合目录和呈现 通过一个简单易用的应用程序，同时进行分析和基因组浏览，将它们提供给临床研究人员 能力，GenVision Ultra。该提案的重点是增强我们高度可扩展的XNG组装管道 具有全自动检测和注释来自单倍型的所有类别的变体所需的工具 相控序列。对核心XNG组件的新颖改编将划分与来自 这些可能代表用于并行处理的SV(目标1)。匹配的读数将与参考对齐 使用XNG，而假定包含SV的读数将使用我们的Long Read从头组装和注释 汇编程序(LRA)。基于参考的比对将使用一种新的贝叶斯分类器进行分阶段，以产生两个 SNV/Small Indel调用和注释之前的单倍型序列(目标2)。短文打磨整个 组装将按需提供。完整的小型变种和SV配置文件以及基础 最终用户可以在GenVision Ultra中访问装配数据。此外，应用程序将具有离散的 筛选和统计工具，用于识别单个样本中感兴趣的基因和/或变体 跨队列/人口(目标3)。为了确保该软件满足临床测序市场需求， Arkana实验室已同意提供来自高度精选的HapMap控制的ONT和Illumina序列数据 用他们的肾脏疾病基因板处理过的样本。这些真实世界的数据集与专家一起 Arkana研究人员的解释和反馈为开发 这一关键市场的杰出软件解决方案(目标4)。