桑格测序
艾滋病毒耐药性
全基因组测序
人类免疫缺陷病毒(HIV)
抗药性
医学
大规模并行测序
DNA测序
临床微生物学
计算生物学
病毒学
基因组
病毒载量
生物
遗传学
抗逆转录病毒疗法
微生物学
基因
作者
Frances Jenkins,Thomas Le,Rima Farhat,Angie N. Pinto,Azim Anzari,David Bonsall,Tanya Golubchik,Rory Bowden,Frederick J. Lee,Sebastiaan J. van Hal
摘要
Abstract Detection of HIV drug resistance (HIVDR) is vital to successful anti‐retroviral therapy (ART). HIVDR testing to determine drug‐resistance mutations is routinely performed in Australia to guide ART choice in newly diagnosed people living with HIV or in cases of treatment failure. In 2022, our clinical microbiology laboratory sought to validate a next‐generation sequencing (NGS)‐based HIVDR assay to replace the previous Sanger‐sequencing (SS)‐based ViroSeq. NGS solutions for HIVDR offer higher throughput, lower costs and higher sensitivity for variant detection. We sought to validate the previously described low‐cost probe‐based NGS method (veSEQ‐HIV) for whole‐genome recovery and HIVDR‐testing in a diagnostic setting. veSEQ‐HIV displayed 100% and 98% accuracy in major and minor mutation detection, respectively, and 100% accuracy of subtyping (provided > 1000 mapped reads were obtained). Pairwise comparison exhibited low inter‐and intrarun variability across the whole‐genome (Jaccard index [ J ] = 0.993; J = 0.972) and the Pol gene ( J = 0.999; J = 0.999), respectively. veSEQ‐HIV met all our pre‐set criteria based on WHO recommendations and successfully replaced ViroSeq in our laboratory. Scaling‐down veSEQ‐HIV to a limited batch size and sequencing on Illumina iSeq. 100, allowed easy implementation of the assay into the workflow of a small sequencing laboratory with minimal staff and equipment and the ability to meet clinically relevant test turn‐around times. As HIVDR‐testing moves from SS‐ to NGS‐based methods and new ART drugs come to market (particularly those with targets outside the Pol region), whole‐genome recovery using veSEQ‐HIV provides a robust, cost‐effective and “future‐proof” NGS method for HIVDR‐testing.
科研通智能强力驱动
Strongly Powered by AbleSci AI