The variety of methods and platforms for sequencing genes makes the process easier and harder. It’s easier, because of the simplification of the processes and sophistication of the platforms, but it’s harder, because there are more options to consider. The task at hand and the desired features—such as speed or cost—reveal the best method to use. To make the best choice, getting the opinions of a few experts comes in handy.
At The Sequencing Center, cofounder and head of product and business development Ryan Casey says, “The most common genome-sequencing methods we see are whole-genome sequencing and targeted genome sequencing, specifically for humans.” Although The Sequencing Center focuses on bacterial, yeast, and targeted human sequencing, “when you focus on humans, it’s highly common to see whole-genome sequencing and whole-exome sequencing,” Casey explains.
The key features that determine the best sequencing method—according to Brewster Kingham, director of sequencing and genotyping at the Delaware Biotechnology Institute at the University of Delaware in Newark—are: “desired quality, desired yield, desired read length, and cost.”
In addition to the general methods, the platforms really matter. Shawn Baker, genomics advisor/consultant at SanDiegOmics.com and cofounder of AllSeq.com, says, “There are four main sequencing-platform companies,” and they are: Illumina, Pacific Biosciences, Thermo Fisher Scientific, and Oxford Nanopore Technologies. Baker adds that there are “some specialty platforms” like the Gene Reader from Qiagen for clinical sequencing, and China’s MGI Tech (a subsidiary of BGI) makes platforms primarily for the Chinese market.
Biocompare’s NGS Search Tool
Find, compare and review NGS tools
from different suppliers Search
Specific sequencing platforms offer unique features. “Illumina platforms tend to dominate cost and throughput while having the highest raw-read quality,” Baker says. “PacBio produces long reads and the highest quality consensus sequence. Oxford Nanopore produces incredibly long reads—as well as having a portable form factor. Thermo’s Ion Torrent’s main advantage is a quick runtime.”
With the right method and a matched machine, a scientist can dig lots of data out of all sorts of sequencing.
Picking the process
The application impacts the best sequencing method, but not everyone agrees about the combinations. “We take somewhat of a contrarian view,” Casey says. “When you’re talking about bacteria, whole-genome sequencing is the best approach given that the cost is cheap and the data generated is manageable.” For human research, though, Casey argues that “targeted sequencing is a much better approach than whole-genome sequencing.”
In brief, he points out that whole-genome sequencing is expensive, complex, and provides more information than is needed in most human research. “In reality, for most researchers, they don’t need all 23,000 genes but rather a very small subset” for human research, Casey explains. If scientists select whole-genome sequencing for human research, Casey notes, “they incur significant additional cost, massive amounts of data to sift through, and 99% of it they don’t really need.”
To really find out what technology works the best, it takes experience. At the Delaware Biotechnology Institute, for example, Kingham and his colleagues, he explains, were “trying to elucidate isoforms and splice variants for a set of gene transcripts,” and they “selected PacBio SMRT sequencing because it allowed us to sequence the full length gene transcripts using individual sequencing reads.”
In many cases, even experienced scientists benefit from experts at a sequencing center. For example, Casey and his colleagues did some sequencing for Eli Lily “When we initially talked, they were interested in performing whole-genome sequencing as their concerns were that if they don’t they would be ‘missing out on potentially valuable data,’” Casey says. “The counterpoint we made was that if they go with targeted sequencing, they will be able to get much higher quality data—high resolution—for a fraction of the price.” With a smaller and more manageable data set, Casey explains, “they would be able to hone in on results much quicker, which would allow them to identify additional genes that they could go after in future runs.”
Ultimately, the customer got 4–5 targeted runs for the price of one whole-genome sequencing run. Casey adds that the targeted sequencing “ultimately, accelerated their research. This led us down the path of suggesting a gene panel for HLA Typing provided by Omixon”.
Image: Determining the nucleic acid sequence of DNA—an example shown here—can be done in various ways, and the best one depends on the objectives of an application. Image courtesy of the National Science Foundation.
The size of a project also impacts the best selection for technology. “Essentially all of the ‘population genomics’ studies, such as the UK 100,000 Genomes Project and the NIH All of Us research project, are using the larger Illumina platforms HiSeq X or NovaSeq due to the tremendous throughput and cost savings compared with other platforms,” Baker explains. “Oxford Nanopore’s handheld MinION is rapidly becoming the sequencer of choice when working on infectious diseases, like Zika and Ebola, due to its portability and rapid turnaround time—critical aspects when working in the field.”
In some cases, commercially available sequencing accessories for specific applications must be tested on different platforms. As an example, August Woerner, an assistant research professor at the University of North Texas Health Science Center in Ft. Worth, and his colleagues studied mitochondrial DNA (mtDNA) for forensic applications with Thermo Fisher Scientific’s Precision ID mtDNA Whole Genome Panel and compared the results from running the panel on Thermo Fisher’s Ion S5 System and Illumina’s MiSeq FGx Desktop Sequencer.
In a 2018 issue of Forensic Science International: Genetics, Woerner and his colleagues reported: “This study demonstrates that the Ion and MiSeq platforms provide consistent haplotype estimation of the whole mitochondrial genome, thus providing further support for the reliability and validity of the Precision ID mtDNA Whole Genome Panel.”
So from single-celled organisms to humans, what scientists want to learn from sequencing—as well as how much information will be processed and how much can be spent—indicates the best way to go about it and with which technology. Not all experts will agree, but their opinions help researchers get started. As Casey concludes: “Ultimately, the context of the research—in terms of the scope of what their field of study is—determines what type of sequencing they should realistically go for.”