Skip to main content
Premium Trial:

Request an Annual Quote

Vendors Say Array-based Sequence Capture Still Alive, But Barely


By Justin Petrone

Call it sequence capture or target enrichment, most researchers today are using solution-phase technology to perform targeted next-generation sequencing. Still, there is demand for array-based sequence capture from clients conducting smaller-scale studies or those who want to try the approach before moving on to larger projects, according to several sources familiar with the market.

Fred Ernani, marketing director for Agilent Technologies' SureSelect target enrichment platform, said recently that the firm's customers continue to purchase capture arrays from the company, though most prefer Agilent's solution-phase offering.

"Generally, those who are using [array-based capture] are looking at very small numbers of samples, and they want to revise their design multiple times," said Ernani. "It has proven to be a very effective method for these very small-scale studies," he said. Ernani added that some customers like to work with capture arrays before "jumping to in-solution" for bigger studies. "It is just much more efficient from a workflow standpoint," Ernani said of the latter method.

Ernani declined to discuss what proportion of SureSelect users rely on array-based target enrichment versus solution phase. But Mike Leous, marketing manager for arrays at Roche NimbleGen, said that comparing "glass to solution, the ratio we see purchased is about 1 percent to 99 percent."

Michael Jansen, an application scientist at Roche NimbleGen, said that there are some advantages to ordering the firm's sequence capture arrays over its solution-phase SeqCap EZ offering. "You can order custom arrays in order sizes of one," Jansen said. "When you do it in solution, you have to order at least 12 reactions," he said. "So people who want to try the system and want to do very small sample sizes use array-based" capture.

Ernani, Leous, and Jansen all spoke to BioArray News at the American Society of Human Genetics annual conference in Montreal earlier this month. During the conference, Roche NimbleGen introduced a pre-capture multiplex target enrichment protocol to enable multiple DNA samples to be barcoded and captured in a single SeqCap EZ Library reaction for exome or custom capture experiments. The protocol, yet to be launched, will become available with the third version of Roche NimbleGen's SeqCap EZ Exome Library. Agilent at the same conference released the fourth version of its SureSelect Human All Exome Kit.

Roche NimbleGen, Agilent, RainDance Technologies, Illumina, Fluidigm, and other firms that compete in the target-enrichment market have developed numerous solution-phase offerings over the past few years. Of larger vendors, only Agilent and Roche NimbleGen offer array-based sequence capture, and their respective capture arrays made their debut before any solution-phase products were commercially available, with Roche NimbleGen entering the market in the second half of 2008, followed closely by Agilent in early 2009 (BAN 2/10/2009).

Earlier this year, Agilent said it won the first patent covering array-based sequence capture, though the impact of the awarded IP was not immediately clear (BAN 3/15/2011). Ernani declined to discuss the IP environment.

Jansen said that Roche NimbleGen is currently focused on developing solution-phase sequence capture products for its customers. "The advantage of the solution phase is that you can actually buy this in 96-well plates and you can run this through a robot and automate it," Jansen said, "so most customers running high-throughput studies will use in-solution" capture. At the same time, he said that all of the firm's new capture products can be run on either arrays or in solution.

"Everything that you get in solution, you can put on an array; it doesn't matter," Jansen said. "If something is new in solution phase, and you like the product but would like it on a glass slide, it can be done."

Ernani also said that it was "theoretically possible" to move all of the firm's solution-phase products to arrays.

Echoing Agilent's Ernani, Jansen said that Roche NimbleGen's customers sometimes use both approaches. "It is not uncommon for customers to order arrays and then move to solution phase as they scale up, and if they have small projects, they stick with arrays," Jansen said. But in addition to throughput considerations, array-based sequence capture has other limitations, he noted.

[ pagebreak ]

"The disadvantage for someone new is that the arrays require certain equipment such as the hybridization station," said Jansen. "You are looking at at least $20,000," he said. "Someone who is going to start solution phase will not have to buy any additional equipment, other than regular lab equipment such as a thermal cycler and pipettes and a heat block and so on," Jansen added. "Those who are using other array products and have the equipment and have smaller projects that they want to try out, they go for arrays."

Roche NimbleGen's Leous noted that the company's capture probes are manufactured on arrays using its maskless array synthesis technology and then cleaved off for use in solution.

According to Ernani, Agilent's baits are printed onto glass in its array fabrication facility and then cleaved off of the glass. They then are taken through a "complex and highly quality-controlled process by our bioreagents manufacturing facility in Cedar Creek, Texas, and kitted up there."

One factor that could lead to more demand for capture arrays is the availability of lower-throughput benchtop sequencers, like Illumina's MiSeq or Life Technologies' Ion Torrent PGM. Ernani said that Agilent is "well positioned" to serve such customers, but predicted that "a lot of desktop sequencing will still likely involve the use of the same design over and over" and argued that the company's in-solution product is "better suited" for such studies.

Pros and Cons

One researcher who continues to use capture arrays is Patrick Schnable, director of the Center for Plant Genomics at Iowa State University in Ames. BioArray News has spoken to Schnable several times about his use of arrays and next-gen sequencing in maize-related projects (BAN 4/26/2011, BAN 1/26/2010.

"We continue to use array-based sequence capture and are in fact this week ordering an array for an experiment that will involve the capture of a few dozen genotypes," Schnable said this week. "We are also planning another array-based experiment that will involve the capture of a few hundred individuals," he said.

According to Schnable, the "pros" of array-based sequence capture versus solution phase include a "set-up cost per experiment that is less for small- to medium-sized experiments, as we can order a single array." He also maintained that "set-up time is less" and said that arrays are "delivered substantially faster than liquid-phase probe sets."

Thirdly, for many labs, an array-based system yields "more on-target reads," though he cautioned that "results may vary, especially if you are new to this."

One "con" of array-based sequence capture is that it requires some equipment, "but we already have it and many core facilities probably have it already too," said Schnable. Array-based capture requires "more technique," he added, referring to technical expertise, but said that his lab as "already worked out the protocols."

Ultimately, he said that both systems work for researchers in that they allow similar multiplexing of libraries prior to capture "If I were going to capture thousands of genotypes with the same probes I would use liquid phase and a robot," said Schnable, "but for the dozens or hundreds of genotypes we typically analyze, array-based technology remains our choice."

Daniel Turner, director of applications at Oxford Nanopore Technologies, said that he used Agilent capture arrays before he joined the firm from the Wellcome Trust Sanger Institute last year. An early-access Agilent customer, Turner said that even before he left the Sanger, "there had been a definite shift to solution capture."

He described the advantages of solution-phase capture as ease of use, noting that it is "possible for one person to perform many more capture experiments per day." He also said that solution capture can be automated easily, requires a lower mass of DNA, and offers higher sensitivity and specificity compared with array-based capture.

"It could be that array capture has improved since then, but I think the main reason for the [performance] differences … is that on array, you need a massive excess of template over probes to drive the capture," Turner told BioArray News this week. "Any inefficiency in the library prep will affect the results badly," he said. "In solution, it's possible to use an excess of probes over template to drive the hybridization, so it is less vulnerable to fluctuations in library quality."

Have topics you'd like to see covered in BioArray News? Contact the editor at jpetrone [at] genomeweb [.] com