Skip to main content
Premium Trial:

Request an Annual Quote

China's Direct Genomics Unveils New Targeted NGS System Based on Helicos Tech for Clinical Use


NEW YORK (GenomeWeb) – Chinese startup Direct Genomics is aiming to tap into China's clinical sequencing market with a targeted single-molecule sequencing system that is based on former Helicos BioSciences' technology.

The company launched its platform, dubbed GenoCare, at the Advances in Genome Sequencing Technology conference in Shenzhen, China, this week and plans to start beta-testing the system next year.

Direct Genomics Founder and CEO Jiankui He told GenomeWeb that the company licensed single-molecule sequencing patents, previously licensed by Helicos, from the California Institute of Technology. The firm is also working with the China Food and Drug Administration to bring the system through regulatory clearance, as well as with three early-access hospitals in China: Shenzhen Women's and Children’s Health Hospital, Shenzhen People's Hospital, and South Medical University.

In addition, Direct Genomics described a prototype system in a proof-of-concept publication posted to the pre-print server BioRxiv this week, demonstrating the sequencing of eight known mutations in the EGFR, KRAS, and BRAF cancer genes.

"It's the first single-molecule sequencing for the clinic that has been designed in China," Michael Deem, chair of the department of bioengineering at Rice University and scientific advisor to Direct Genomics, told GenomeWeb.

The GenoCare system will make use of Helicos' chemistry and reagents, but in a new instrument, Stephen Quake, a professor of bioengineering at Stanford University who will serve on Direct Genomics' scientific advisory board, told GenomeWeb in an email. Quake, a co-founder of Helicos, was not involved in the development of the system, but He was a former postdoc in his lab.

Deem, also an author on the BioRxiv paper, added that the instrument will be much smaller than Helicos' HeliScope system, about the size of a "large file cabinet," he said. The system is around 1.4 meters (4.6 feet) tall and half a meter (1.6 feet) wide, about one-fourth the size of the original HeliScope, He said.

The GenoCare system will have a throughput of less than 10 gigabases per run, He said. Accuracy will be "approaching 100 percent" for a sample sequenced to 5x coverage, and at launch, the system will have read lengths of around 35 bases. Turnaround time will vary depending on the application, but would likely be around 15 hours for a 50-gene cancer panel and down to four hours for more targeted applications.

He declined to disclose specifics on pricing, but said that the instrument would be around 30 percent less expensive than competitors' systems.

In the BioRxiv paper, the researchers targeted eight known mutations in the EGFR, KRAS, and BRAF genes, including six point mutations and two short deletions, that are related to drug response. They designed eight capture probes upstream of each drug response-related mutation and tested the sequencing on two sets of synthetic DNA sequences — one that had the wildtype variants and one with the mutations.

They sequenced the DNA to an average of 1,954-fold coverage, but noted that coverage depth varied among the different DNA templates, possibly because "hybridization efficiency for DNA templates is sequence-dependent and the secondary structures that involve the target region can also affect hybridization efficiency."

Read lengths in the paper were just 10 bases, but He noted that in the several months since, read lengths have increased to 19 bases and will soon be up to 35 bases. Moving to a four-color chemistry as well as improvements to the hardware would help increase read lengths further to eventually 140 bases.

The researchers wrote in the paper that the consensus sequence accuracy was 100 percent. To evaluate how coverage depth impacted accuracy, the team subsampled reads. At 1x coverage, average accuracy was 95 percent, and at 5x coverage the "consensus accuracy is approaching 100 percent," the team noted. In addition, they found that the average substitution error rate was around 0.52 percent per base.

Next, the researchers created mixtures of wildtype and mutant DNA to establish the lower limit of detection for mutant alleles. They found they were able to detect mutant sequences at 3 percent frequency.

One key to the GenoCare technology, which was also part of Helicos' core technology, is the use of total internal reflection fluorescence (TIRF) microscopy to increase the signal-to-noise ratio, He said.

In TIRF, a laser is angled at the glass surface, which causes it to reflect downward rather than through the glass, which generates a short-lived "evanescent wavefront" that illuminates only the contents on the immediate surface of the flow cell where the captured DNA is located, and not the unincorporated oligonucleotides and other fluids in the flow cell, increasing the signal-to-noise ratio. That fluorescence from the labeled DNA is then detected via microscopy and an electron-multiplying CCD camera.

Also similar to Helicos' technology is that the GenoCare does not require any sample prep machines. The capture step takes place directly on the flow cell and is automated, a feature He said is especially suited for a hospital setting.

While the current GenoCare prototype uses similar chemistry and reagents as Helicos' HeliScope instrument, one change that Direct Genomics is working to implement is the use of four differently colored dyes, which He said would improve the accuracy, since each base would have a different colored fluorescent label. The current chemistry uses only one dye.

The electron-multiplying CCD camera is another difference between GenoCare and the HeliScope instrument. While the HeliScope also used a CCD camera, "the technology itself has been improved and is much better than it was 10 years ago," He said.

The flow cell configuration will also be slightly different. Direct Genomics plans to offer both 16- and 32-channel microfluidic chips, as opposed to the HeliScope's 25-channel flow cell.

One patient sample will run in one channel of the chip, He said, and it will be possible to run just a single sample on the chip without using extra reagents.

"Every sample is independently controlled by the microfluidics," He said.

Currently, Direct Genomics is running the GenoCare internally for its hospital collaborators, but He said that the company would be installing beta systems at these sites in the second half of 2016, followed by a broader commercial launch in early 2017.

The company plans to focus on the Chinese market, in particular the 1,400 or so cancer-focused hospitals in China, Deem said, and does not yet have plans to enter the European or US markets.

Researchers at the three beta-testing hospitals are evaluating the use of the system for sequencing viral DNA and circulating tumor DNA to inform treatment of hepatitis B and cancer, respectively. The South Medical University team is looking to use GenoCare to identify drug resistance mutations in patients with hepatitis B — a huge problem in China, which has "one-third of the world's infected population," Wang Zhanhui, professor of infectious diseases at South Medical University, said in a statement.

He said that Direct Genomics is also working on developing a 50-gene cancer panel, as well as tests for noninvasive prenatal screening.

Deem said that compared to competing technologies, one main advantage of GenoCare is that it will be much simpler to use, particularly in terms of sample prep, which will essentially consist solely of a DNA fragmentation step prior to loading DNA onto the flow cell. "That reduces the cost and chances of sample handling error," Deem said.

Nonetheless, Direct Genomics will still face significant competition, especially because the clinical sequencing market in China has taken off, recently, with the approval of NGS instruments and noninvasive prenatal tests by the CFDA.

Direct Genomics will likely have to compete with those already approved tests and instruments, including Berry Genomics, with a cleared NIPT on the NextSeq CN500, and BGI with two cleared NIPTs on the BGISEQ-100 and BGISEQ-1000, which are based on Thermo Fisher Ion Torrent and Complete Genomics technology, respectively. In addition, BGI announced this week that it had launched a desktop sequencer based on Complete Genomics technology, BGISEQ-500, which it plans to bring through CFDA clearance.