Skip to main content
Premium Trial:

Request an Annual Quote

Variantyx Combines Long Nanopore, Short Illumina Reads for Rare Disease Whole-Genome Diagnostic Test

Premium

NEW YORK – Genetic testing firm Variantyx has launched a rare disease diagnostic test that combines short- and long-read whole-genome sequencing and promises to deliver more comprehensive variant analysis to help improve diagnostic yield.

Dubbed Genomic Unity 2.0, the test integrates Illumina short-read data with long reads from Oxford Nanopore Technologies' sequencing technology.

"We are trying to improve the number of variants that we can detect using both short- and long-read sequencing," said Christine Stanley, Variantyx's chief director of clinical genomics. "The overall mission is to improve the diagnostic yield and to provide diagnoses for patients that are going through long diagnostic odysseys."

Prior to Genomic Unity 2.0, Framingham, Massachusetts-based Variantyx had been offering the standard Genomic Unity test, which is primarily based on short-read sequencing. While that assay can also deploy long-read sequencing as a secondary tool to confirm certain variants, Stanley said, the annotation is done solely using short-read data, leaving out some variants that are not seen by the technology. In contrast, Genomic Unity 2.0 places equal emphasis on short-read and long-read sequencing to achieve the most comprehensive variant analysis. 

Typically, the Genomic Unity 2.0 assay starts with 5 ml of patient blood, from which DNA is extracted using automated platforms with the same protocol for both sequencing modalities. After quality control, the sample bifurcates to different library prep protocols for short-read and long-read sequencing.

Lucy Kaplun, Variantyx's director of advanced development, said the Genomic Unity 2.0 assay aims to achieve 30X coverage for the short reads, which are generated on the Illumina NovaSeq X Plus system. For nanopore sequencing, the test shoots for 12X to 15X converge using the Oxford Nanopore PromethIon 24 device, where multiple samples are loaded onto the same flow cell.

After basecalling, the short-read and long-read data are integrated for joint variant annotation using Variantyx's bioinformatics platform. The variants then undergo clinical interpretation and further evaluation by a clinician, based on the patient's disease phenotype.

Subsequently, the results are passed on to a variant scientist and compiled into a report, which is further reviewed and signed off by a clinical geneticist before being reported back to the ordering physician.

According to Variantyx's website, Genomic Unity 2.0 analyzes single nucleotide variants, deletions, insertions, intronic, regulatory, and intergenic variants. It also includes structural variants such as copy number variants, duplications, and deletions. Furthermore, the nanopore data enables methylation profiling for diagnosing Angelman, Prader-Willi, and fragile X syndrome, as well as variant phasing, all unique features of the test.

Stanley said Genomic Unity 2.0 targets patients "who want the best and most comprehensive test available for diagnosis at any age." The assay can be ordered as a first-line test, or clinicians can order the standard Unity test first and upgrade to version 2.0 if the results are negative. In that case, the sample will typically undergo additional DNA extraction and long-read sequencing, after which both the short-read and long-read data will be reanalyzed together.

Currently, the turnaround time for Genomic Unity 2.0 is 10 to 12 weeks, on par with the standard Genomic Unity test, though it is often shorter, according to Stanley.

Institutional pricing for the test is $4,400 for the proband and $1,000 for each comparator, adding up to $6,400 for a parent-child trio. The cost for reflexing to 2.0 after the standard Unity test is about $1,900.

"We are trying to submit whatever we can to insurance to make the amount that we are asking from the patient as low as we can," a Variantyx spokesperson said. The company already has a Proprietary Laboratory Analyses (PLA) code for the original Genomic Unity test, Stanley noted, and is working on obtaining a PLA code for the 2.0 test.

Stanley said Variantyx decided to adopt Oxford Nanopore sequencing instead of Pacific Biosciences' HiFi sequencing mainly because of the former's lower upfront capital investment and sequencing costs. "It's probably four times more expensive to do PacBio," she noted. "We can do the dual [sequencing] for cheaper than just doing the long reads on PacBio. So, it makes more sense to go with Oxford Nanopore."

Still, incorporating nanopore sequencing into rare disease diagnostic testing that conforms to CLIA regulations and CAP requirements was not a trivial effort for Variantyx, which claims to be the first company to do so. The firm not only needed to show that the technology works but also to demonstrate its performance using real clinical samples, which can be "quite challenging" to source, Stanley said.

"Long-read sequencing has [so far] mostly been used as a research tool," Kaplun noted. "We have real people that are waiting for diagnosis. … We need to have very stable technologies that perform superbly every time."

One challenge of applying nanopore sequencing in a diagnostic setting, she said, was to lock the protocols, especially given the fast-evolving nature of the technology. "We revalidate everything when we upgrade [the workflow]," Stanley said. "There is a myriad of steps that we have to take when we upgrade anything. So, it is definitely a challenge."

Additionally, Kaplun said that while nanopore sequencing enables "very decent detection" of single nucleotide variants, the technology's error profile for indels, especially in homopolymeric regions, can still be "problematic," though it is constantly improving.

While long-read sequencing holds the potential to replace other assays for genetic testing, Stanely noted, that goal is still far away given the technology's cost and accuracy, and a combined test such as Genomic Unity 2.0 still has a place in the current diagnostic ecosystem. "There is the idea that you can get all this information off a single [long-read] platform," she said. "Right now, we are not there yet."