Skip to main content
Premium Trial:

Request an Annual Quote

CUHK Team Sequences Plasma DNA on MinIon, Demonstrating Feasibility of NIPT

Premium

NEW YORK (GenomeWeb) – The research team from the Chinese University of Hong Kong that developed the technology used in Sequenom's noninvasive prenatal screening test has now shown that Oxford Nanopore's MinIon device can sequence DNA from maternal plasma, demonstrating its potential for noninvasive prenatal testing.

Reporting their work in a letter in the journal Clinical Chemistry this month, the researchers sequenced plasma samples on the MinIon from pregnant women in their third trimester, non-pregnant women, and men.

Dennis Lo, senior author of the study and director of the Li Ka Shing Institute of Health Sciences, told GenomeWeb that his group has been testing the MinIon since April 2014. He said that although additional improvements need to be made before nanopore sequencing could be used clinically for NIPT, it has potential.

"The relative simplicity and potential rapidity of nanopore sequencing would, with further development of the technology, allow the potential for near-patient molecular testing," Lo said in an email. A patient could go to a prenatal clinic and have NIPT results in just a few hours, he said.

In addition, he said the lower cost of the MinIon compared to other sequencing instruments could help "expand the reach of NIPT to many more parts of the world."

In the feasibility study, the researchers collected plasma from women pregnant with male fetuses, women pregnant with female fetuses, non-pregnant women, and men. They collected plasma from pregnant women in their third trimester, so the levels of circulating cell-free fetal DNA would be higher. From each group, they pooled plasma so that they had around 20 ml for each cohort.

Next, they concentrated the plasma DNA down to 85 microliters and prepared DNA libraries following the standard protocol for MinIon library preparation. In the future, however, Lo said increasing the amount of adaptors might help improve efficiency of library prep. The MinIon library prep was optimized for long pieces of DNA, but cell-free DNA found in plasma tends to be highly fragmented. With the same amount of DNA, shorter fragments will have more ends that need adaptors ligated than longer DNA, so increasing the amount of adaptors available could potentially help speed the process up, he said.

The majority of plasma DNA fragments were in the 155 bp to 168 bp range, there were a few fragments longer than 1,000 bp, with the longest reaching 5,776 bp.

Lo said that long plasma DNA fragments have not been detected on other NGS instruments, although researchers using gel electrophoresis have previously seen long plasma DNA fragments. "The ability of the MinIon to sequence such long fragments is very exciting," he said, and opens the door for research on the "biology and diagnostic impact of such long circulating DNA molecules."

After sequencing, the researchers used Oxford Nanopore's cloud-based base calling software, operated through its Metrichor architecture, and the 2D reads were extracted and aligned to the reference genome.

Sequencing took between six and 24 hours per library, and between 26.9 percent and 32.5 percent of reads passed the basecaller. On average, 82.7 percent of bases aligned to a matching base in the reference. From the 2D reads, 16.9 percent aligned to a unique genomic location and were analyzed further.

Next, the researchers calculated the distribution of reads that aligned to each chromosome and found that the distribution to the autosomes was "comparable to those expected for the mappable human genome," the authors wrote.

As expected, they found that the proportion of reads that mapped to the X chromosome was lower from the male plasma sample than the female plasma sample, at 2.70 percent and 5.22 percent, respectively. In addition, no reads from the non-pregnant female plasma samples mapped to the Y chromosome.

Approximately .30 percent of reads from the male plasma sample aligned to the Y chromosome, while .11 percent of reads from the female with a male fetus sample aligned to the Y chromosome. The women carrying female fetuses also generated a small fraction of reads, .018 percent, that aligned to the Y chromosome, but the authors wrote that that was "consistent with previous data." From the plasma pool of women carrying male fetuses, there were around 1 percent fewer reads aligning to the X chromosome than from the women carrying female fetuses.

"Fetal DNA sequences and chromosome X dosage differences between male and female fetuses are detectable by nanopore sequencing," the authors wrote, suggesting the "potential feasibility of using nanopore sequencing for the noninvasive detection of fetal chromosomal aneuploidies."

The authors did not, however, analyze whether they could detect copy number differences between autosomes, since they had pooled plasma samples from multiple individuals. Lo said that the same Z-score method for detecting aneuploidy from other NGS instruments could be used with the MinIon, although he said the group would use a different alignment platform that is optimized for long reads.

One major limitation of using the MinIon for NIPT is that a high concentration of DNA is needed in a small volume. In the feasibility study, the researchers achieved this by pooling samples to reach 20 ml of plasma.

Lo said that an advantage of nanopore sequencing is that no amplification is done, which reduces bias. But, that means that there is subsequently less DNA. To ensure that the DNA molecules reach the nanopore, there needs to be a high concentration of DNA in a small volume. Future improvements to DNA extraction and the library prep process could allow for this, he said.

Another current limitation is the error rate of the MinIon, which Lo said would have to be reduced to a level comparable with existing NGS platforms before it could be used for NIPT. Finally, he said he would like to see base calling enabled in a client's own laboratory, rather than always having to be done in the cloud, which would speed up the process.

Higher throughput is also "an important parameter for NIPT," and as such, Lo said that he would be interested in testing Oxford's PromethIon instrument when it becomes available.

Aside from NIPT, Lo's team recently began a project to evaluate nanopore sequencing for detecting circulating tumor DNA, although he said that project is still in the very early stages.