A research group at the Navy Medical Research Center in Rockville, Md., is using two 454 Life Sciences’ FLX sequencers as part of an effort to develop pathogen-detection assays and improved vaccines and therapeutics for biodefense purposes.
Timothy Read, head of the genomics group at the NMRC’s Biological Defense Research Directorate, told In Sequence in a recent interview that his group began using the 454 technology to sequence microbial genomes in early 2005, making it one of the earliest adopters of the technology.
Read, who left the Institute for Genomic Research in 2003 to found the genomics group at the NMRC, said that his lab was initially conceived as a functional genomics group, “and it still is, but we got the opportunity to get involved with the new next-generation sequencing technologies and, basically, we got in very early and we started thinking about how we could apply these technologies to biodefense problems of detection.”
Like other research groups who have adopted the 454 platform, Read noted that the technology is “perfectly suited to microbial de novo genome sequencing.”
The NMRC researchers are using the sequencers to create what Read described as “digital strain collections,” which are in silico catalogs of existing bacterial strain collections. Read said he expects these collections to help speed up early screening for detection and diagnostic development, as well as vaccine and therapeutic design.
His team has already studied Burkholderia isolates, strains of Bacillus cereus, B. thuringiensis, B. mycoides, B. weihenstephanesis, and B. pseudomycoides, and relatives of Yersinia pestis.
The particular aim of this effort is to assemble a large database of closely related species in order to help researchers better distinguish pathogenic strains from non-pathogenic stains and background noise.
“We are looking to basically take a snapshot of what we know at the moment exists in nature,” he said. “In most cases, [these organisms] are non-pathogenic, but they can confound detection assays, so the more information we have about the sequence of close relatives, the better we will be able to design detection assays with fewer false positives.”
The digital strain collection is also expected to help researchers develop vaccine targets with “a broader host range than just one pathogen, because we will be able to have information about a pathogen and all its close relatives and see if we can look for vaccine targets that may potentially work against all these strains,” he said.
The digital strain collection could also be useful in metagenomic studies, he said, because those projects require more data about individual genomic strains in order to identify individual species with large environmental samples.
“[Because] we are building up databases where we know what the organism is, we can probably help metagenomic work, especially soil [studies], because we are sequencing a lot of soil organisms,” he said.
Read said that the scale of this effort requires the throughput of next-generation sequencing methods. “We were getting on the way with Sanger technologies, but the potential is to sequence hundreds and thousands of closely related bugs, rather than tens. And I think that’s the direction that we are getting to very rapidly,” he said.
The NMRC group recently completed sequencing 15 Burkholderia isolates and released the data to the National Center for Biotechnology Information last week, Read said.
The group has also sequenced 10 Bacillus cereus strains alongside five B. thuringiensis, two B. mycoides, one B. weihenstephanesis, and one B. pseudomycoides strains. In an abstract describing this project, Read and colleagues write that they achieved 20-fold to 35-fold sequence coverage for each genome.
In another project, the NMRC team sequenced eight relatives of Yersinia pestis that are commonly found in the environment and “offer baseline data for biothreat detection assay design and analysis of the evolution of virulence of pathogenic strains,” according to the project’s abstract. The researchers achieved 25-fold to 42-fold sequence read coverage for these genomes.
Read said that the NMRC team and its collaborators are currently preparing manuscripts for several of these projects that they hope to submit for publication before the end of the year. “We are not particularly interested in publishing these genomes one by one. We are interested in multi-genome papers,” he said. “We are just waiting to finish large sets of data, and then we will look towards writing them up for publication.”
“We got the opportunity to get involved with the new next-generation sequencing technologies and, basically, we got in very early and we started thinking about how we could apply these technologies to biodefense problems of detection.”
It currently takes about a day to sequence each genome, Read said, noting that a large portion of his group’s time is “spent acquiring the strains rather than actually creating the data.”
Another challenge is bioinformatics. Read said that the lab is working with the BioTeam, a bioinformatics consulting firm, to ensure that its IT infrastructure is scalable. In addition, the group is using an automated protein functional analysis pipeline called PIPA developed at the Department of Defense’s Biotechnology High-Performance Computing Software Applications Institute.
Currently, he said, “we can comfortably handle about 50 to 100 genome analyses at a time.”
The NMRC lab has two FLX instruments that it acquired with an undisclosed amount of funding from the Defense Threat Reduction Agency. The lab also has an Affymetrix array system, a NimbleGen array platform, a MAUI hybridization system from BioMicro, and a Beckman FLX robot.
Read said that he is also keeping an eye on other next-gen sequencing platforms, although the group has no definite plans to acquire another system at this time.
“We obviously think that there are different applications for these different technologies, and we are very interested in them,” he said. “In terms of generating whole-genome sequences, it may be that we could combine 454 with another technology to get better coverage and hopefully … optimize the sequence data quality. So we will be looking at that as these technologies become available.”