Sierra Roberto, Roch Mélanie, Moraz Milo, Prados Julien, Vuilleumier Nicolas, Emonet Stéphane, Andrey Diego O
Infectious Diseases Division, Department of Medicine, Geneva University Hospitals and Faculty of Medicine, 1205 Geneva, Switzerland.
Department of Microbiology and Molecular Medicine, Faculty of Medicine, University of Geneva, 1206 Geneva, Switzerland.
Pathogens. 2024 Aug 28;13(9):730. doi: 10.3390/pathogens13090730.
In the context of increasing antimicrobial resistance (AMR), whole-genome sequencing (WGS) of bacteria is considered a highly accurate and comprehensive surveillance method for detecting and tracking the spread of resistant pathogens. Two primary sequencing technologies exist: short-read sequencing (50-300 base pairs) and long-read sequencing (thousands of base pairs). The former, based on Illumina sequencing platforms (ISPs), provides extensive coverage and high accuracy for detecting single nucleotide polymorphisms (SNPs) and small insertions/deletions, but is limited by its read length. The latter, based on platforms such as Oxford Nanopore Technologies (ONT), enables the assembly of genomes, particularly those with repetitive regions and structural variants, although its accuracy has historically been lower.
We performed a head-to-head comparison of these techniques to sequence the VS17 isolate, focusing on resistance gene alleles in the context of a surveillance program. Discrepancies between the ISP ( allele identified) and ONT ( and alleles identified) were observed. Conjugation assays and Sanger sequencing, used as the gold standard, confirmed the validity of ONT results. This study demonstrates the importance of long-read or hybrid assemblies for accurate carbapenemase resistance gene identification and highlights the limitations of short reads in the context of gene duplications or multiple alleles.
In this proof-of-concept study, we conclude that recent long-read sequencing technology may outperform standard short-read sequencing for the accurate identification of carbapenemase alleles. Such information is crucial given the rising prevalence of strains producing multiple carbapenemases, especially as WGS is increasingly used for epidemiological surveillance and infection control.
在抗菌药物耐药性（AMR）不断增加的背景下，细菌全基因组测序（WGS）被认为是检测和追踪耐药病原体传播的一种高度准确且全面的监测方法。存在两种主要的测序技术：短读长测序（50 - 300个碱基对）和长读长测序（数千个碱基对）。前者基于Illumina测序平台（ISP），在检测单核苷酸多态性（SNP）和小插入/缺失方面具有广泛的覆盖范围和高准确性，但受其读长限制。后者基于牛津纳米孔技术（ONT）等平台，能够组装基因组，特别是那些具有重复区域和结构变异的基因组，尽管其准确性在历史上一直较低。
我们对这些技术进行了直接比较，以对VS17分离株进行测序，重点关注监测计划背景下的耐药基因等位基因。观察到ISP（鉴定出的等位基因）和ONT（鉴定出的 和 等位基因）之间存在差异。作为金标准的接合试验和桑格测序证实了ONT结果的有效性。本研究证明了长读长或混合组装对于准确鉴定碳青霉烯酶耐药基因的重要性，并突出了在基因重复或多个等位基因情况下短读长的局限性。
在这项概念验证研究中，我们得出结论，近期的长读长测序技术在准确鉴定碳青霉烯酶等位基因方面可能优于标准的短读长测序。鉴于产生多种碳青霉烯酶的菌株患病率不断上升，此类信息至关重要，尤其是随着WGS越来越多地用于流行病学监测和感染控制。
