Life Technologies announced this week that it has established a collaborative effort involving researchers from several public health organizations and other organizations to evaluate and adapt an Ion Torrent PGM protocol to sequence the influenza genome for flu tracking and vaccine development.
Researchers affiliated with the effort, called the Global Influenza Network, have been working with the PGM for about half a year, Dan Didier, Life Technologies' public health director, told Clinical Sequencing News. So far, they have completed their evaluation of a protocol for influenza A typing and defined standard operating procedures.
"They've actually also analyzed a lot of the information and are sharing it between the groups … [but] they are still finalizing the bioinformatics pathway before we make it available for a larger group to start using this," he said.
The collaboration is a pilot program to evaluate the efficacy of influenza virus typing using Life Tech's whole-genome influenza sequencing protocol for the PGM. According to Life Tech, the group has already determined that the approach for influenza A is accurate (100 percent concordant with capillary methods, according to Didier), sensitive, and economical — at a cost of about $100 per isolate.
The company said in a statement that current sequencing costs limit the amount of sequencing that can be performed on samples of patients infected with the flu. These samples are sequenced each year by public health agencies to learn about flu subtypes circulating in particular regions and to determine the strains used to design a vaccine against the following year's epidemic. However, according to Life Tech, only about 20 percent of patient samples are currently sequenced.
The plan is for the network to publish its results some time later this year, and for Life Tech to then more widely publicize the protocol the group has optimized. The hope is that the PGM's turnaround time will allow more rapid and comprehensive flu virus typing than with the Sanger-based methods that are currently in use at most public health labs.
According to Didier, the choice to work with a network of outside researchers to test and demonstrate the approach means that it has more immediate applicability than if Life Tech kept it in house.
"We're hoping to get that out as soon as possible. I don't think it would affect this flu season but maybe next season," Didier said. "The goal here, if you look at what is normally done with only about 20 percent of isolates being sequenced, the goal is to get that to 100 percent."
As part of the pilot, members of the network have been evaluating the approach against their standard workflows, which include methods such as Sanger sequencing and other next-gen sequencing platforms.
Last October at Life Tech's Ion Torrent User Group Meeting, one of the network members, Steve Glavas of the Swedish Institute for Communicable Disease Control, gave a presentation on his team's work testing the PGM approach.
In his presentation, which is available on video, Glavas said his institute has relied on Sanger sequencing as a clinically robust method, but that it is not rapid enough to be used in a pandemic situation. "Sanger gives us the data we need but we can't scale up in case of a pandemic. It's impossible," he said.
As part of the Life Tech influenza network, Glavas said his group has developed a PGM workflow, starting with automated library preparation using a LabchipGX and ending with automated assembly, built using a CLC Bio assembly algorithm.
According to Glavas, this allows the group to run 26 influenza genomes per day. Considering that his team only looks at about 400 genomes per season, that's "just a couple days of work," he said in his presentation.
Glavas said his team did an initial experiment to test the PGM protocol using Life Tech's FluAmp kit using 20 influenza isolates — 10 on a 314 chip and 10 on a 316 chip. Overall, he said the result was "very even coverage."
For the 314 chip, he said the group has been able to get between 52x coverage, which is "good enough" for them, and 710x coverage ("way overkill") over the range of RNA fragments. With the 316 chip, he said the team saw roughly 10 times that amount of data.
Initially comparing its reads from the isolates against older Sanger data, Glavas said the group saw several differences in the sequences, but discovered that they were not sequencing errors on the part of the PGM, but rather differences in the DNA due to using isolates instead of clinical samples.
In a further comparison the group saw no sequence deviation between the two methods, and overall, Glavas said his team has seen about a one percent error rate for its entire workflow from extraction to analysis.
David Wentworth, director of viral programs at the J. Craig Venter Institute, represents another team in the network partnership. He told Clinical Sequencing News that his group currently does next-gen flu sequencing using a combination of Illumina machines and the Roche 454.
He said the hope in adopting the PGM would be for it to replace the 454 in that scheme.
"Part of the reason we want it, and that we are in this group, is for really rapid turnaround of high-priority viruses," Wentworth said. "The hope was that since we would man it directly in my lab and not in a core we could say something is high priority and get it done right away."
According to Wentworth, the institutes involved in the network are working with each other now to tweak the protocol to get the best genome coverage they can of the influenza virus.
"It's complicated in some ways because of the segmented nature of the genome," he said. "We use PCR at the front end to amplify the segments, and inherently, the small segments amplify better than the long segments."
Additionally, he said his team is also experimenting with phosphorylation to help with issues with ends of fragments.
"We think we've figured out that we could be helped by phosphorylation of the 5' end … but we are still trying to figure out how we can improve that, experimenting with phosphorylated primers versus phosphorylating after the reaction," he explained.
"But getting better coverage, that's the main thing we are all talking about" in the network, he said.
Wentworth added that his group has not worked with Life Tech's PathAmp Flu kit yet, which may also help with amplification issues.
According to Life Tech's Didier, the company expects the participating groups to be closer to publishing data from their work with the PGM after February.