NEW YORK (GenomeWeb) – Genia has published a proof of principle study of its nanopore-based sequencing-by-synthesis technology, although the final commercial product will look substantially different and be suitable for clinical diagnostics, according to Genia CEO Stefan Roever.
Roever did not provide a timeline for when the firm, which Roche acquired in 2014, would launch an instrument. In 2013, he had said that Genia planned to launch by the end of 2014, but after the acquisition, Roche "revisited what they thought the specs would need to be to launch a sequencer into the clinical sequencing market," Roever told GenomeWeb. Roche had a "substantially more stringent set of requirements" than the targets Genia had previously set as goals for launching a viable product, he said.
In the new study, which was published this week in the Proceedings of the National Academy of the Sciences, Genia researchers, in collaboration with Jingyue Ju's lab at Columbia University, George Church's group at Harvard Medical School, and researchers from the National Institute of Standards and Technology, demonstrated that they could design a 264-nanopore array and use a tagging technique to discriminate the four nucleotides and sequence DNA.
The approach Genia is taking relies on tag-based sequencing-by-synthesis. The firm has designed four unique oligonucleotide tags that are coupled to the terminal phosphate of each of the four nucleotides. In addition, they attached a highly processive DNA polymerase to each alpha-hemolysin nanopore. When a tagged nucleotide forms a ternary complex with the polymerase and the primer/template DNA, the tag enters the nanopore, producing a specific ion current blockage that enables the researchers to record that nucleotide.
Genia last presented data at the end of 2013 and in a poster at the Advances in Genome Biology and Technology meeting in 2014, prior to being acquired by Roche.
Roever said that the recent PNAS study uses the same chemistry and construct as work that was previously shown. The paper, however, provides a few more details about how the data was generated.
In the PNAS study, the researchers demonstrated their technique on an 83-base template strand of DNA as well as on 12-base homopolymeric regions.
They were able to determine the correct sequence orders for a 4-base sequence, a 20-base strand, and a 12-base homopolymer, although senior author Ju told GenomeWeb that they have since produced reads 1 kilobase in length.
In addition, they observed a so-called "stuttering," where one nucleotide tag appears to be captured in the pore several times in a row. Such stuttering is caused by the tag being repeatedly bound to and released from the polymerase. Although one potential advantage of such stuttering is that it would provide multiple signals for the same base, which might improve the accuracy, it would complicate the sequencing of homopolymers, Ju said.
The researchers found that they were able to get rid of the stuttering by changing the buffer conditions and showed that they could read a 12-base homopolymer stretch.
Genia's approach is different from Oxford Nanopore Technologies' strand sequencing method, which relies on distinguishing differences in the current elicited by a set of DNA bases in the pore. Instead, the Genia system relies on detection of the tag, and requires a polymerase, but has single-base resolution.
Ju said one advantage of using tags is that they can be designed in such a way that different tags result in very different current signatures — greater than the current differences between the four natural nucleotides — yielding more contrast for sequence determination.
In addition, the tagged DNA molecules sit within the pore for between 17 and 30 milliseconds, depending on the tag, according to the study.
Ju said that since the PNAS study, the researchers have continued to optimize the tags with respect to size, charge, and structure, and tag optimization has been a major focus of the Columbia group's work. Since publishing a paper describing the general Nano-Tag technique in Scientific Reports in 2012, the researchers have made the tags larger and negatively charged, which "allow more optimal buffer conditions for [the] polymerase reaction," Ju said.
Roever added that the technology is now well beyond what was demonstrated in the PNAS study, although he declined to provide specifics. "In terms of the read length, accuracy, yield, and speed, there's been improvement in all those major metrics," he said.
Aside from improving on the metrics, Roever said, the group has also made changes to the technology, although the underlying principle is the same.
Going from what was demonstrated in the PNAS paper to a commercial product has been a long road, he said, especially working towards Roche's goal of "having a sequencer to be used for clinical purpose."
After acquiring Genia, Roche made substantial investments in the company, growing it to 150 employees from 30, Roever said.
When the system launches, it will compete with Oxford Nanopore's MinIon, which is being adopted especially for microbial and viral sequencing. Nick Loman's lab at the University of Birmingham, for example, showed that it could be used in the field to monitor the Ebola outbreak in real-time and Charles Chiu's group at the University of California, San Francisco, is developing a pathogen detection pipeline for the MinIon, to eventually be used in diagnostics.
But it is unclear how the Genia system will fit into Roche's overall sequencing landscape. Roever said that the system would have applications that "support clinical diagnostics." However, Roche has an agreement with Pacific Biosciences to develop a clinical sequencing platform, and if Roche decides to commercialize the Genia technology for diagnostic applications, it would lose its exclusive right to distribute PacBio's technology for diagnostics.