我国宏基因组高通量测序检测血液微生物游离DNA的方法及质量保证情况分析

Analysis of the methods and quality assurance of metagenomic next-generation sequencing to detect the microbial cfDNA from blood samples in China

目的:调查我国实验室开展血液标本微生物细胞游离DNA(mcfDNA)宏基因组高通量测序(mNGS)的检测方法及质量保证情况。方法:2020年10月向来自全国的80家实验室发放了mNGS检测血液标本中mcfDNA的调查问卷。问卷内容包括mNGS分析前、分析中、分析后及性能确认开展情况4个部分。(1)分析前:对标本质量的要求,如对标本的采集、储存及运输条件等;(2)分析中:mcfDNA的提取流程、文库的质量要求、测序平台的使用及生物信息学分析软件等;(3)分析后:对mNGS的结果解释标准;(4)性能确认开展情况:对各类病原体的最低检出限。要求各实验室依据实际情况填写调查问卷。对上述调查问卷的回报结果进行统计分析。结果:80家实验室包括20家医疗单位和60家独立医学实验室。80.0%(64/80)的mNGS实验室检测血液中的mcfDNA时表示血浆及血清标本均可使用,其余实验室(16/80,20.0%)只采用血浆标本。参与调查的mNGS实验室所用的测序平台包括illumina 49家(61.3%),华大基因16家(20.0%),Ion Torrent 13家(16.3%),纳米孔测序2家(2.5%)。87.5%(70/80)的实验室使用第三方实验室搭建的集成分析工具,其他实验室(12.5%,10/80)使用开源软件自主搭建了分析平台。实验室间对mNGS结果的解释标准不一,其中标准化后病原体特异性序列数、相对丰度、基因组覆盖率及阴性对照中该微生物的检出情况是实验室考虑的主要因素。大部分实验室(76.3%,61/80)做过mcfDNA测序流程的性能确认实验,各实验室自建mNGS检测流程对革兰阳性菌、革兰阴性菌、真菌、寄生虫及其他病原体的最低检出限主要分布在10~100 拷贝/ml,对DNA病毒的检出限主要分布在500~1 000拷贝/ml。结论:各实验室之间的检测流程具有很大差异,为了确保检测结果的及时准确,需要各实验室积极优化mNGS检测流程,完善质量保证措施,应用于临床前规范开展性能确认工作。

Objective To investigate the methods and quality assurance of metagenomic next-generation sequencing(mNGS)to detect the microbial cfDNA(mcfDNA)from blood samples in different laboratories across China.Methods In October 2020,questionnaires about detecting mcfDNA in blood samples with mNGS were distributed to 80 laboratories across the country.The questionnaire included four parts:pre-analysis,during analysis,post-analysis,and carrying out of performance validation for mNGS.(1)Pre-analysis:the requirements for samples quality,such as collection,storage,the transportation conditions of samples;(2)During analysis:the extraction workflows of mcfDNA,the quality requirements of the library,the application of the sequencing platforms and the bioinformatics analysis pipelines;(3)Post-analysis:the standard of interpretation results for mNGS;(4)Carrying out of performance validation:the minimum detection limit for various pathogens.All laboratories are required to fill in the questionnaire according to the actual situation.The feedback data were summarized and analyzed.Results The 80 laboratories included 20 medical centers and 60 independent medical laboratories.There were 80.0%(64/80)of laboratories indicated that both plasma and serum samples were used to detect mcfDNA in blood,and the rest of the laboratories(16/80,20.0%)only used plasma samples.The sequencing platforms used by mNGS laboratories involved in the survey included illumina(49),Beijing Genomics Institute(16),Ion Torrent(13)and Nanopore sequencing(2).There were 87.5%(70/80)of laboratories used the integrated analysis tools built by the third-party laboratories,and other laboratories(12.5%,10/80)independently built the analysis platform by open-source software.The interpretation criteria of mNGS results varied between laboratories,among which the normalized number of pathogen-specific sequences,relative abundance,genome coverage rate,and the detection of the microorganism in the negative control were the main factors considered by laboratories.Most laboratories(76.3%,61/80)had carried out the performance validation for the mcfDNA mNGS workflows.The limit of detection of the laboratories-developed mNGS workflows for Gram-positive bacteria,Gram-negative bacteria,fungi,parasites,and other pathogens were mainly distributed at 10-100 copies/ml,DNA virus was mainly distributed at 500-1000 copies/ml.Conclusions The mNGS workflows of various laboratories are very different.In order to ensure timely and accurate testing results,every laboratory needs to actively optimize the mNGS testing procedures,improve quality assurance measures,and carry out performance validation before mNGS is widely used in clinical settings.