New method for detecting HIV drug resistance may lead to improved therapies

A team of scientists developed a technology to more accurately detect HIV drug resistance and potentially improve the quality of antiretroviral therapies for people living with HIV. 

The technology is a bioinformatics pipeline, a series of computer programs that process genome sequence data in search of genetic variations indicative of drug resistance. The pipeline, HIVGenoPipe, was developed by scientists in the Applied and Developmental Research Directorate at the Frederick National Laboratory. It is an open-source software freely available for use by scientists and clinicians as described in a recent issue of BMC Bioinformatics. 

Drugs that target proteins of Human Immunodeficiency Virus-1 (HIV-1) will prevent AIDS disease symptoms in affected individuals, keeping them healthy and stopping the virus from spreading. But the virus readily mutates, leading to drug resistance and eroding the effectiveness of treatment. And the virus propagates rapidly, contributing to the ongoing pandemic with more than 1 million new cases a year and more than 600,000 deaths. 

Drug resistance can be reduced by detecting it early and using that information to formulate successful antiretroviral regimens for HIV affected individuals, while also eliminating unnecessary drug use, the scientists said.  

To that end, scientists developed HIVGenoPipe to accurately depict HIV genomic sequence data relevant to the viral proteins targeted by current antiretroviral drugs, proteins such as protease, reverse transcriptase, integrase and those involved in host-virus interactions. The pipeline uses a novel multiple-step hybrid assembly procedure that yields more accurate drug resistance interpretation of the genomic data than other methods. 

HIVGenoPipe was validated against the widely used HyDRA pipeline and standard Sanger sequencing. The validation exercises used a dataset of 30 plasma samples from volunteers living with HIV who had enrolled in clinical trials and were undergoing antiretroviral therapy. 

The consensus sequence data from HIVGenoPipe were more accurate than those from other pipelines because of the multi-step hybrid assembly procedure that use real-time sample-specific reference sequences, the scientists said. And HIVGenoPipe matched 99.995 percent of the Sanger sequencing results. 

HIVGenoPipe was developed to work with Illumina MiSeq sequencing, which is commonly used to look for HIV drug resistance because of its low cost and high accuracy. But any Illumina sequencing machine can be used with the pipeline, the scientists said. HIVGenoPipe operates on the Nextflow platform, which is modular, reproducible, integrates with other tools and is portable and scalable. It is designed for intermediate bioinformatics users on private servers in a clinical setting. As currently deployed, however, it is less accessible than web-based applications and may require hardware, software and installation.