NEW YORK (GenomeWeb) – Two research teams from Europe and the US have taken the Oxford Nanopore MinIon sequencer to West Africa to study the Ebola outbreak at its source, allowing them to track down transmission routes quickly with the goal of breaking the chain of infection.
The small size and ruggedness of the MinIon sequencer enabled them to set up shop at their destinations quickly, though the need to upload the raw data for analysis proved challenging for one of the groups.
Both teams, one operating under the umbrella of the European Mobile Laboratory Project (EMLab), the other as part of the US National Institutes of Health, presented early results of their projects at a conference organized by Oxford Nanopore Technologies in London last month, and both plan to publish their studies in the near future.
Miles Carroll, head of research and deputy director of microbiology services for Public Health England, has been involved in providing Ebola diagnostic services in West Africa since March 2014, when he helped set up a mobile laboratory in Guéckédou, Guinea, to provide diagnostic support to Doctors Without Borders. His team is part of the EMLab project, which is funded by the European Commission and now has four diagnostic laboratories in West Africa, which have tested more than 10,000 samples for Ebola.
The diagnostic laboratories use an RT-PCR test on RNA extracted from patient samples to determine whether a patient is infected with Ebola and the viral load in the blood sample. But that test provides no information about the origin of the virus, which only sequencing data can provide.
A subset of 180 Ebola-positive samples, obtained between March 2014 and January 2015, were shipped back to Europe for a retrospective study that involved sequencing the full virus genome to better understand its spread, mutation rate, and adaptation to its host. Results from that project are scheduled to be published in a high-profile journal in the near future, Carroll said.
The phylogenetic trees the researchers obtained from the sequencing data showed how the virus traveled from Guéckédou to Freetown in Sierra Leone before it came back to Guéckédou to cause a second, bigger wave of the outbreak.
"When I was out there a couple of months ago, and I had these data, I thought, 'This is madness, it's all very nice having this one year later, a retrospective look, but what we need is real-time sequence capability,'" Carroll told GenomeWeb. Shipping an established sequencing platform to Guinea was not an option, he said, due to the instrument cost, logistical issues, and the infrastructure and maintenance required for such platforms, although other groups have managed to set up Ebola sequencing labs in Liberia, using the Illumina MiSeq, and in Makeni, Sierra Leone, using Thermo Fisher Scientific's Ion Torrent PGM.
Back in England, Carroll got in touch with Nick Loman at the University of Birmingham, whose lab had independently planned to take Oxford Nanopore's MinIon to West Africa to analyze the Ebola outbreak, and teamed up with his group.
Relative to the size of the outbreak — so far, there have been more than 26,000 cases and 10,000 deaths — little Ebola genome data has been generated, Loman said, even though genomic surveillance data could benefit epidemiologists in the region. "It is important to have that information as soon as possible," he told GenomeWeb.
Taking the MinIon to Africa turned out to be "relatively easy compared to taking the larger instruments," according to Loman. His graduate student, Joshua Quick, took off for Guinea in mid-April, with all instrumentation and reagents contained in a single piece of checked luggage.
The largest piece of equipment was a PCR machine, followed by four laptops, three MinIon devices, a small centrifuge, a heat block, and two polystyrene boxes with refrigerated and frozen consumables, including the MinIon flow cells, magnetic beads, library prep kits, and enzymes.
Quick set up the equipment in a laboratory at Donka Hospital in Conakry, which is affiliated with the EMLab project. The lab had no equipment other than a fridge and a freezer, as well as back-up power from a generator. Within a few days, Quick generated Ebola sequence data from samples shipped in from Coyah, a few hours' drive north, where the outbreak is still ongoing.
"We quickly learned from that data that the strain that had infected somebody in Coyah was not actually coming in from Freetown or Conakry; it was coming in a roundabout way from Guéckédou, which was a big surprise to us," Carroll said.
Over the next couple of weeks, Quick sequenced Ebola from more than a dozen samples in total. Loman's group in Birmingham analyzed the data, building phylogenetic trees, which revealed that the source of the outbreak in Coyah was actually two lineages —one from Guéckédou, the other from Freetown.
"We were able to do a whole array of things that would never have been possible," Carroll said, including informing the authorities of the source of infection for a number of Ebola patients. "They could not work out the transmission chain, and this sequencing has been able to do that," he said.
Since Quick went back to England, the MinIon sequencing lab has been transferred from Donka Hospital in Conakry to a mobile EMLab in Coya, where it produces Ebola sequencing data for every new case on a daily basis. The data are analyzed in Birmingham, and phylogenetic trees go back directly to epidemiologists from the World Health Organization and the Guinean authorities.
"It's a frontline service that we are now providing to the WHO and the Guinean coordination committee" to track down the transmission routes of the final cases, Carroll said.
Among other things, the lab recently started to sequence Ebola from semen samples of recovered patients, he said, because it looks like the virus can survive in seminal fluid for up to six months post infection. The lab also sequences samples from newly infected cases that had sexual contact with recovered patients in order to show that this is how they became infected.
According to Loman, the biggest challenge with using the MinIon in Guinea has been the availability of a stable Internet connection to upload the raw sequence data in order to perform base and variant calling, and to transmit the sequence data to the UK for the phylogenetic analysis.
Each sample generates about 400 megabytes of data, he said, which are "quite easy" to send over a 3G mobile phone network, but sometimes the connection switches back to slower networks that cannot cope. His team also explored using satellite Internet but found that to be both expensive and slower than a 3G network.
In the absence of a stable Internet connection, it would be helpful to be able to perform base calling on a local computer, Loman said.
Another challenge is power. The MinIon itself can run off the power of a laptop battery, and each sequencing run only takes about an hour, so power outages are not much of a problem. In contrast, the PCR machine requires continuous power, Loman said, though battery-operated PCR machines could be used in the future, or alternatives to PCR amplification could be employed.
The sequencing lab receives RNA extracted from Ebola-positive samples from the diagnostic lab. Ebola has a 20-kilobase single-stranded RNA genome, which the lab PCR-amplifies in 10 sections of roughly 2-kilobase length. The 10 amplicons are then pooled in equal amounts and prepared for sequencing using the standard MinIon library prep.
Currently, the lab runs one sample per flow cell but plans to multiplex several barcoded samples in the future, Loman said.
The sequencing lab is kept separate from the diagnostic lab, Carroll said, to prevent cross-contamination with PCR products.
One advantage of larger sequencing platforms, which provide a higher throughput than the MinIon, Carroll said, is that they can be used to sequence not only Ebola virus but also the RNA of the host, in order to study the host response. "The Minion won't replace the big sequencing platforms for research back in a nice Western laboratory, but definitely on the front line, the MinIon has a whole spectrum of advantages over the classical sequencing platforms."
Oxford Nanopore provided support for the project, Carroll said, donating both equipment and consumables and providing technical support.
Loman said he and his collaborators are planning to release data from the Ebola project on a website soon, prior to publishing the study in a scientific journal.
Another Ebola MinIon study conducted in West Africa by NIH researchers is currently under review at a journal. Thomas Hoenen, a visiting postdoc in the Laboratory of Virology at the Rocky Mountain Laboratories of the NIH in Hamilton, Montana, who presented results from the study at Oxford Nanopore's meeting, told GenomeWeb that he cannot provide any information at this time because of the journal's embargo policy.
He said he sequenced Ebola samples for about three weeks in February of this year, working at the CDC/NIH diagnostic laboratory at the Eternal Love Winning Africa (ELWA) campus in Monrovia, Liberia. Like the EMLab team, he found uploading the raw data over a mobile phone network to be a limiting factor, so part of his analysis was conducted after his return to the US.