Skip to main content
Premium Trial:

Request an Annual Quote

Wellcome Sanger Benchmarking Study Highlights Suitable Enzymes for NGS Library Amplification


NEW YORK – Researchers from the Wellcome Sanger Institute have benchmarked over 20 high-fidelity enzymes for next-generation sequencing library amplification, shedding light on the best PCR polymerases and products for both short-read and long-read applications.

Of the roughly two dozen enzymes tested, they found that the Quantabio RepliQa HiFi ToughMix, the Watchmaker Genomics Equinox Amplification Master Mix, and the Takara Ex Premier DNA polymerase were the "enzymes of choice" for amplifying short-read Illumina libraries. Meanwhile, Quantabio’s RepliQa also trumped other candidates for long-read library amplification.

"The motivation for doing the study was to let the community know the useful enzymes to use in their [NGS] experiments," said Michael Quail, a principal scientist for sequencing R&D at the Wellcome Sanger Institute and the first author of an April Microbial Genomics paper that described the benchmarking results.

According to Quail, the recent analysis was a sequel to a previous benchmarking effort also done by his group a decade ago to determine the best-performing enzyme for short-read Illumina library amplification. That study, which was published in Nature Methods then, identified Roche’s Kapa HiFi DNA polymerase as the frontrunner.

"Companies have developed new enzymes over the past decade," Quail said. "We wanted to repeat that study with some of the newer enzymes that have come out."

For the current study, the researchers compared over 20 commercially available high-fidelity PCR enzyme products from a slew of NGS sample prep reagent providers, such as Roche, Agilent Technologies, New England Biolabs, Bio-Rad Laboratories, Qiagen, and Thermo Fisher Scientific, just to name a few.

As part of the experiment design, Quail said, his team had talked to "all the major vendors" and asked them to recommend their best enzymes and the associated optimal reaction conditions to be incorporated into the study. 

To benchmark the enzymes for short-read amplifications, the researchers prepared Illumina adapter-ligated libraries as PCR templates from a set of four microbial genomes with varying GC content: Bordetella pertussis, Escherichia coli, Clostridioides difficile, and Plasmodium falciparum.

For each polymerase tested, the researchers amplified 1 ng of templates from each genome using reaction conditions recommended by manufacturers for 14 PCR cycles, which aimed to "exacerbate any biases" of the candidate enzymes, according to the study.

The PCR products were subsequently cleaned up and size selected using AmPure beads and sequenced on an Illumina NovaSeq 6000 platform after a series of QC measures.

The sequencing results were then aligned to the reference genomes, serving as the basis for several evaluation metrics, including GC bias, coverage of the genome, and the uniformity of the coverage.

Those enzymes that performed well during the initial assessment proceeded to a second round, which included differing cycling conditions as well as two additional products that were previously unavailable — the Watchmaker Equinox Amplification Master Mix and the Takara Ex Premier DNA polymerase.

Overall, the researchers said they observed "distinct differences" among the tested enzymes in terms of yield and the evenness of amplification, reinforcing the notion that PCR performance can potentially skew sequencing results.

After two rounds of bake-offs, they considered the performance of Quantabio RepliQa, Watchmaker Equinox, and Takara Ex Premier to be "impressive," delivering uniform coverage of the libraries with extreme GC content that, according to Quail, was "almost as good as" PCR-free data.

The Sanger researchers further evaluated these top-performing enzymes with the library generated with human genomic DNA from the genome in a bottle sample. The results showed that all three products performed "remarkably well," with RepliQa standing out with the highest amplification precision for single nucleotide polymorphisms and indels.

Besides short reads, the study also sought to identify the best enzymes for long-range PCR, which is often used for targeted sample prep or to compensate for low library input in long-read sequencing. To do that, they prepared size-fractionated, adapter-ligated PCR templates with the yeast genome, which were then deployed to test the performance of over a dozen polymerases, using manufacturer-recommended protocols and Pacific Biosciences HiFi sequencing.

"Some of those enzymes just didn't work," Quail said, adding that for those that did, the team evaluated their yield and coverage, especially given that long-range PCR tends to preferentially amplify smaller library templates.

In the end, the results indicated that Quantabio’s RepliQa and Takara’s Terra PCR direct polymerase were the best performing enzymes for long reads and achieved "near complete genome coverage" of the library, according to the study authors. The former also had fewer substitution errors, making it the winner overall.

Still, Quail pointed out that even though RepliQa ranked at the top, the enzyme showed some preferential amplification of smaller fragments. As such, he also thinks that the long-range PCR enzymes still have a way to go in general compared to their short-read counterparts.

Although the long-read benchmarking was done using PacBio sequencing, Quail also noted that he would expect "very similar results" with the Oxford Nanopore Technologies platform.

"Quantabio is proud to see our RepliQa HiFi ToughMix selected as the enzyme of choice for NGS library amplification in this important benchmarking study," Brian Komorous, head of marketing, applications, and support at Quantabio, wrote in an email. "We will continue to apply our enzymology expertise to provide scientists around the world with the most robust DNA and RNA amplification reagents to help unlock future discoveries."

"We were impressed with the scope and depth" of the study, Trey Foskett, cofounder and CEO of Watchmaker, wrote in an email, adding that the company was "pleased" to see its Equinox polymerase outperformed many others across various metrics.

Meanwhile, a spokesperson for Qiagen, whose product did not perform well enough during the initial short-read benchmarking to make it to the second round, said the company "suspect[s] that the UCP HiFidelity PCR mix, which is mentioned [in this study], was not actually the product tested."

"This stems from the fact that due to production constraints at the time the Sanger Institute initiated their research and reached out to Qiagen, the UCP HiFidelity PCR mix was not available," the spokesperson wrote in an email. As a result, the only HiFi product available for customers was UCP’s precursor, the HotStart HiFidelity polymerase kit, which was "not designed for NGS and was discontinued already some time ago."

Quail, on the other hand, said to the best of his knowledge and as backed by order records, his team did receive and use UCP for this study. "Our priority is to ensure the accuracy and integrity of our research findings," he responded. "To provide additional assurance, we will work together with Qiagen to verify the results of this experiment and will provide a comprehensive response as soon as possible."

"It's a very powerful and very insightful study," said Xinkun Wang, director of the NUSeq core facility at Northwestern University, who was not involved in the study. According to Wang, his lab has previously evaluated Watchmaker and Takara sample prep products, which have generated good results. Therefore, the Sanger Institute study, which he considers to be "well designed," reinforces their good impressions of these companies’ chemistries.

On the other hand, Wang said he was a little surprised to see some legacy vendors well-known for their enzymes, such as NEB, did not make it to the top of the list. "It is kind of surprising, but I still have good faith in NEB products in general," he noted.

Nonetheless, Wang thinks the data from the Sanger researchers will be informative for his own decision-making one day. Take Quantabio RepliQa, for example. "It looks like their enzyme is very robust, in terms of both short-read sequencing and long-read sequencing," Wang said. "If we need to evaluate new kits [in the future], that product will definitely be on the top of our list."