复旦大学附属儿科医院高通量测序数据一体化全流程闭环分析系统及临床应用案例分析

Next generation sequencing data analysis pipeline of Children's Hospital of Fudan University and its clinical application

背景目前的临床遗传诊断中,外显子组捕获测序(ES)和全基因组测序(WGS)均有广泛的应用场景且在综合性价比和变异检测范围等方面各有优势;建立支持两种不同建库策略测序方案的整合性遗传诊断流程,是进一步提高遗传诊断敏感性和检测效率的关键。目的整合ES以及WGS场景下的多变异类型分析、遗传病复杂临床表型的标准化与结构化、表型导向的遗传变异分析系统,建立从临床检测申请到遗传报告反馈的一体化流程。设计流程开发。方法(1)建立复旦大学附属儿科医院高通量测序数据一体化全流程闭环分析系统(复旦流程3.0)各个模块:①病史处理,②结构化遗传表型,③测序实验,④变异检测,⑤变异解读,⑥质控复核,⑦基因型表型联合分析。(2)代表性病例重测分析:根据结论性变异的类型和诊断难度选择代表性病例,展示代表性病例从病史处理开始到报告草稿为止各模块的运行情况。主要结局指标代表性病例分析过程中的结构化表型、数据质控、变异检测及解读、最终诊断。结果对3例代表性病例的重分析中,分别进行了经济版家系WGS、先证者WGS以及ES;病史成功提取结构化表型;FastQ及BAM文件结果数据质控良好;检测到的变异经解读之后进行基因型表型联合分析,分别检测出3例病例的复杂遗传模式。例1:ANTXR2基因点突变NM_058172:c.1294C>T与4q21.22约13 kb的结构变异缺失形成隐性纯合致病;例2:线粒体MT-ND6基因致病变异m.14459G>A,异质性>99.5%;例3:SMN1基因纯合致病变异NM_000344:c.863G>T合并SMN1基因单拷贝缺失。结论复旦流程3.0可整合ES和WGS数据,处理不同类型的变异结果,最终形成遗传诊断,且对复杂变异类型有高效处理能力。

Background Currently in clinical genetic diagnosis,both exome capture sequencing(ES)and whole genome sequencing(WGS)have a wide range of application scenarios.Each has its own advantages in terms of either better cost-effective performance or a wider variant detection range.The establishment of an integrative genetic diagnosis process that supports two different library preparation and sequencing protocols is essential to further improve the sensitivity and efficiency of genetic testing.Objective By integrating the analysis of various variant types fitting both ES and WGS scenarios,the normalization and structuring of complex clinical phenotypes of genetic diseases,and the phenotype-oriented genetic variation analysis system to establish an integrated process from the application of genetic test to the feedback of a diagnostic report.Design Process development.Methods An integrated full-process closed-loop analysis system for high-throughput sequencing data(Fudan Process 3.0)was established including the modules of processing the medical history,extracting structured terms of phenotype,sequencing experiment,detecting variants,interpretating variants,checking quality control,and analyzing both genotypes and phenotypes.In terms of test-retest analysis of representative cases,we selected representative cases with various type of conclusive pathogenic variants and diagnosis difficulties to present the analysis process from sequencing experiments and clinical history to the generation of a draft report.Main outcome measures The structured phenotype terms of patients,the data quality control parameters,the status of variant detection and interpretation,and the final diagnosis during the analysis of representative cases.Results During the re-analysis of 3 representative cases,the optimized trio genome sequencing,proband-only WGS and CES were carried out respectively.The structured phenotype was successfully extracted from the medical history.The data quality of FastQ and BAM files was well controlled.After interpretation,a combined genotype and phenotype analysis was performed to detect the complex inheritance pattern of three cases respectively.In example 1,detected point mutation NM_058172(c.1294C>T and 4q21.22 about 13 kb structural variant deletion on ANTXR2)matched the recessive inheritance model.In example 2,a pathogenic variant m.14459G>A on mitochondrial gene MT-ND6 with heterogeneity>99.5%was detected.In example 3,a homozygous pathogenic variant NM_000344(c.863G>T combined with a single-copy deletion of SMN1 gene)was detected.Conclusion The Fudan Process 3.0 is well functioned in processing either ES or WGS data to analyze various variant types and draw genetic diagnosis conclusions,especially in handling cases with complex variant types.