Methods for improving clinical diagnostic by detection, prediction, interpretation and prioritization of aberrant transcriptome variations

通过异常转录组变异的检测、预测、解释和优先排序来改进临床诊断的方法

• 批准号: 10451556
• 负责人: Yoseph Barash
• 金额: $ 33.99万
• 依托单位: UNIVERSITY OF PENNSYLVANIA
• 依托单位国家: 美国
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-08-03 至 2024-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10451556
• 关键词: Address; Adopted; Age; Algorithms; Alternative Splicing; Benchmarking; Cells; Child; Clinical; Code; Complex; Computers; DNA; Data Analyses; Data Set; Defect; Detection; Development; Diagnosis; Diagnostic; Disease; Event; Excision; Exhibits; Exons; Gender; Genes; Genetic; Genetic Diseases; Genetic Variation; Genomics; Genotype-Tissue Expression Project; Goals; Human; International; Letters; Libraries; Manuals; Medical; Mendelian disorder; Messenger RNA; Methods; Modeling; Molecular Diagnosis; Mutation; Normal tissue morphology; Outcome; Pathogenicity; Patients; Pediatric Hospitals; Performance; Phenotype; Philadelphia; Positioning Attribute; Program Research Project Grants; Protein Isoforms; Proteins; RNA; RNA Processing; RNA Splicing; Regulation; Reverse Transcriptase Polymerase Chain Reaction; Sampling; Site; Technology; Testing; Tissues; Training; United States National Institutes of Health; Untranslated RNA; Variant; Work; causal variant; clinical application; clinical diagnosis; clinical diagnostics; cohort; computational pipelines; congenital anomaly; deep learning model; exome; exome sequencing; follow-up; genetic testing; genetic variant; heterogenous data; improved; interest; novel; prediction algorithm; predictive modeling; programs; standard of care; tool; transcriptome; transcriptome sequencing

The goal of this research program project is to develop methods to improve clinical diagnosis of children with rare Mendelian disorders. Even with the most advanced standard-of-care genetic test of exome sequencing (ES) diagnostic rate is still below 50%. One reason for this rate is that mutations in non-protein coding regions or those that are synonymous (code for the same protein) are generally discarded even though these could be deleterious due to their effect on the processing of RNA transcribed from the underlying gene. We propose 2 complementary methods to help improve clinical diagnosis: The first is “RNA-first”, where our algorithms suggest which clinically accessible tissue (CAT) to use for RNA sequencing, then compare the results to a larger pool of donors to detect which RNA processing variations may be deleterious. The second is a “DNA-first” approach where we develop “RNA splicing code” models that predict the effect of genetic variations on RNA processing in a given tissue of interest. The two approaches, “RNA-first” and “DNA-first”, will be combined into a clinical diagnostic pipeline at the Children Hospital of Philadelphia (CHOP) and applied to solve undiagnosed cases at CHOP and other centers, including the NIH’s Undiagnosed Disease Network (UDN).

该研究计划项目的目的是开发方法来改善患有儿童的临床诊断 罕见的孟德尔疾病。即使有最先进的外显子组测序标准基因测试（ES） 诊断率仍然低于50％。此速率的原因之一是非蛋白质编码区域的突变或 那些是代名词（同一蛋白质的代码）通常被丢弃，即使这些可能是 由于它们对从基础基因转录的RNA处理的影响而有害。我们建议2 完全有助于改善临床诊断的方法：第一个是“ RNA-First”，我们的算法建议 哪些临床可访问的组织（CAT）用于RNA测序，然后将结果与较大的库进行比较 供体检测哪些RNA处理变化可能会被删除。第二个是“ DNA-First”方法 在开发“ RNA剪接代码”模型的地方，可以预测遗传变异对RNA处理的影响 给定感兴趣的组织。两种方法是“ RNA-First”和“ DNA-First”，将合并为临床 费城儿童医院（CHOP）的诊断管道，并申请解决未诊断的病例 CHOP和其他中心，包括NIH未诊断的疾病网络（UDN）。