Researchers from several institutions have developed a high-resolution melting assay that can provide measurements of HIV diversity that are highly concordant with equivalent sequencing-based measurements at a fraction of the cost.
The group believes that the HRM approach, which uses melt-curve analysis technology from Idaho Technology, offers a promising way to assess common measures of the diversity and complexity of HIV infections in cases where full sequence data is unnecessary and time and cost are considerations.
The group, led by researchers at Johns Hopkins University, has been working on the HRM approach for several years. In their most recent publication, released online in the Journal of Clinical Microbiology last week, the researchers measured whether diversity scores from the HRM assay could predict three diversity measures previously obtained from next-gen sequencing data — percent diversity, percent complexity, and a measure called Shannon entropy.
Overall, the group found that the HRM approach's quantitative HIV diversity scores were "highly associated" with sequence-based diversity measures when testing two different areas of the HIV genome.
In addition, both methods comparably measured an increase in diversity over time, the group reported.
And while the results of the two methods were comparable, "equipment, labor, reagent, and supply costs for analysis using the HRM diversity assay are substantially lower than the costs associated with analysis using NGS," the authors wrote.
Mathew Cousins, a John Hopkins PhD candidate working on the HRM approach and the first author of the study, told PCR Insider this week that the group sees two main target applications for the assay — using viral diversity as a way to measure disease severity or predict clinical outcomes, and using it as a way to measure disease duration.
"No one has seriously proposed using something as expensive as next-gen sequencing to [establish disease] incidence," Cousins said. So this assay really gives us the potential to measure [HIV] diversity where I don't think there was a potential before."
The group's HRM assay measures the range over which amplified HIV DNA melts to give a numeric diversity score.
"When you amplify [the DNA pool]," Cousins explained, "you expand the population, and if you have lots of different variants, after you amplify them, you make a bunch of heteroduplexes — each of which is going to have a slightly different melting temperature."
"The sum total of those DNA variants will lead to a broadening of the temperature over which the amplicon melts," he said.
According to Cousins, the HRM assay's performance in comparison with NGS-based diversity measures was very strong. "It is good support for the method, and for the application of the method we are targeting," he said.
In the study, the researchers used a subset of data from the Rakai Community Cohort Study in Uganda in which investigators had already analyzed HIV genomes with Roche 454 sequencing. The Hopkins-led team selected a cohort of 120 paired longitudinal samples from 110 individuals to test the HRM assay and looked at two regions of the HIV genome for each sample, HIV GAG and HIV ENV.
Cousins explained that there are hundreds of different diversity measures that are commonly calculated from sequence data. The group chose three: percent diversity, percent complexity, and "Shannon entropy" — a measure that incorporates both the number of distinct reads in a sequencing alignment and the proportional representation of those reads.
The researchers used the HRM assay's diversity scores as a predictor of the various diversity measures calculated from the sequence data, measuring the HRM method's ability to predict the sequencing results.
For all three, and for both the genome regions, he said the group found a very strong association between the sequence-based measures and the HRM assay's diversity score.
"We don't necessarily hope for a particular result," he said. "But we were happy to see the strong relationship."
Of course there are numerous questions that the simplified HRM approach can't answer, Cousins said. For example, detailed sequence information would be necessary to compare the HIV populations of two individuals for information on infection linkage.
For detailed information about superinfections — cases of patients who are infected multiple times — detailed mutational information is often also important, he said.
But, Cousins said, though sequence data "gives you a fine detail, [researchers] often then go back and align the reads and mathematically generate these diversity measures."
"If you're going to end up reporting diversity as a single measure, and … you don't need the sequence data to build trees or look at relationships, why not use the cheaper assay and avoid the cost and the hassle? That's kind of our idea and our approach. We were trying to come up with a way to circumvent that," he said.
While the researchers state in the paper that the HRM assay costs "substantially" less than the NGS-based assay, they don't provide a detailed cost comparison of the two approaches.
Cousins said the group has already tested the HRM assay as a method for determining HIV incidence and published a proof of principle describing this early work in PLoS One last November.
According to Cousins, this early work hinted that looking at more regions of the HIV genome might help maximize the utility of the HRM assay to measure incidence. The group is now studying a large number of samples targeting eight different regions, and plans to use the study to statistically determine how well the method functions as an incidence assay.
At the same time, the researchers are continuing to mine the Rakai cohort to demonstrate the assay's potential for their second target application, Cousins said. "We are looking at how the diversity we measured relates to the clinical outcome of these individuals — their time to AIDS or time to death," he explained.
The group also recently submitted a paper better describing how different mutations control the melting curve peak width that informs the HRM assay's diversity scores.
In the JCM paper, the researchers report the HRM assay approach might also be useful for evaluating diversity in other pathogens. Cousins said the group would most likely target another quickly evolving virus next. "[Hepatitis C virus] would be an obvious choice," he said.