Ssekagiri Alfred, Jjingo Daudi, Lujumba Ibra, Bbosa Nicholas, Bugembe Daniel L, Kateete David P, Jordan I King, Kaleebu Pontiano, Ssemwanga Deogratius
Department of General Virology, Uganda Virus Research Institute, Entebbe 31405, Uganda.
Department of Immunology and Molecular Biology, Makerere University, Kampala 10206, Uganda.
Bioinform Adv. 2022 Nov 28;2(1):vbac089. doi: 10.1093/bioadv/vbac089. eCollection 2022.
Next-generation sequencing (NGS) enables reliable detection of resistance mutations in minority variants of human immunodeficiency virus type 1 (HIV-1). There is paucity of evidence for the association of minority resistance to treatment failure, and this requires evaluation. However, the tools for analyzing HIV-1 drug resistance (HIVDR) testing data are mostly web-based which requires uploading data to webservers. This is a challenge for laboratories with internet connectivity issues and instances with restricted data transfer across networks. We present QuasiFlow, a pipeline for reproducible analysis of NGS-based HIVDR testing data across different computing environments. Since QuasiFlow entirely depends on command-line tools and a local copy of the reference database, it eliminates challenges associated with uploading HIV-1 NGS data onto webservers. The pipeline takes raw sequence reads in FASTQ format as input and generates a user-friendly report in PDF/HTML format. The drug resistance scores obtained using QuasiFlow were 100% and 99.12% identical to those obtained using web-based HIVdb program and HyDRA web respectively at a mutation detection threshold of 20%.
QuasiFlow and corresponding documentation are publicly available at https://github.com/AlfredUg/QuasiFlow. The pipeline is implemented in Nextflow and requires regular updating of the Stanford HIV drug resistance interpretation algorithm.
Supplementary data are available at online.
下一代测序（NGS）能够可靠地检测1型人类免疫缺陷病毒（HIV-1）少数变异体中的耐药突变。关于少数耐药性与治疗失败之间关联的证据不足，这需要进行评估。然而，用于分析HIV-1耐药性（HIVDR）检测数据的工具大多基于网络，这需要将数据上传到网络服务器。对于存在网络连接问题的实验室以及网络数据传输受限的情况而言，这是一项挑战。我们提出了QuasiFlow，这是一种用于在不同计算环境中对基于NGS的HIVDR检测数据进行可重复分析的流程。由于QuasiFlow完全依赖于命令行工具和参考数据库的本地副本，它消除了将HIV-1 NGS数据上传到网络服务器所带来的挑战。该流程将FASTQ格式的原始序列读数作为输入，并生成PDF/HTML格式的用户友好报告。在20%的突变检测阈值下，使用QuasiFlow获得的耐药性评分分别与使用基于网络的HIVdb程序和HyDRA网络获得的评分100%和99.12%相同。
QuasiFlow及相应文档可在https://github.com/AlfredUg/QuasiFlow上公开获取。该流程在Nextflow中实现，并且需要定期更新斯坦福HIV耐药性解释算法。
补充数据可在网上获取。
