Traditional sequencing is often likened to making a smoothie: researchers blend a bunch of cells, obtain an average sequence, and draw conclusions on the ingredients that comprise the slush. More recently, scientists have gained the ability to perform single-cell sequencing, which can reveal rare variations between cells and the evolution of cell lineages. But current methods require reading the genome in short sections and therefore often fail to capture complex repetitive regions, which scientists are increasingly linking to health and disease. Long-read technologies could overcome this pitfall; however, their methods require much more DNA than can be extracted from a single cell. Single-cell, long-read sequencing has remained frustratingly out of reach.
That is, until now. By combining “two very innovative approaches”—a cutting-edge DNA amplification technique with the latest advances in DNA sequencing—a team of scientists have applied long-read technology to single cells, says Alexander Hoischen, a researcher of genomic technologies at Radboud University Medical Center in The Netherlands who was not involved in the research. “This was unthinkable just two or three years ago,” says Hoischen.
The feat may allow for a more detailed look at mutations underlying all sorts of diseases, experts tell The Scientist.
This was unthinkable just two or three years ago.—Alexander Hoischen, Radboud University Medical Center
Over the past decade, reads—the product of DNA sequencing—have been getting longer. Long-read sequencing has allowed scientists to sequence troublesome “dark regions” of the genome that are inaccessible to short-read technologies, either due to an abundance of guanines and cytosines, or duplicated regions not easily mapped to a chromosome.
However, long-read sequencing requires a ton of DNA. Several micrograms of genetic material are needed, but “a single cell contains just six picograms,” says Joanna Hård, a computational biologist at ETH Zurich in Switzerland. “So substantial amplification is required before you can sequence it” using long-read methods, she says.
And that’s where things get tricky, says Hård, as the primary methods used to amplify DNA are prone to “amplification bias”: the tendency for certain sequences to be ramped up at the expense of others. Now, Hård and colleagues have obtained long reads from individual cells using an improved DNA amplification method. Though not yet peer-reviewed, the results were reported in a preprint uploaded to bioRxiv on January 23.
To minimize amplification bias, the team used a technique called droplet-based multiple displacement amplification. It works by trapping DNA fragments in droplets that contain a limited supply of reagents, preventing over-amplification of certain regions. “There is a more even amplification, so you get better representation of the genome,” Adam Ameur, a bioinformatician at Uppsala University in Sweden, tells The Scientist.
The researchers performed the droplet-based amplification on individual human T-lymphocytes, then generated long reads using PacBio HiFi technology. Compared to short-read sequencing, the new method captured four times as many structural variants—large rearrangements of DNA—including those located in the genome’s inaccessible “dark regions.”
The researchers used the new approach to sequence DNA from two different T-cells obtained from the same person. Their sequencing data revealed 28 somatic mutations that distinguished the two cells, including changes in mitochondrial DNA. The new method could allow scientists to study the impact of somatic mutations in myriad diseases.
For example, tumors often display a mosaic of genetic variation as different cells independently acquire mutations—known as sub-clonal mutations—that make the tumor more aggressive. Improved single-cell DNA sequencing could help “tease out the sub-clonal mutations that often hiding in cancer,” says Christopher Mason, a biophysicist at Weill Cornell Medicine in New York who was not involved in the study.
Healthy cells also accumulate mutations throughout their lives, though this is less alarming in skin and gut cells, which are regularly replenished, than in the long-lived cells in our brain. But somatic mutations in neurons and other brain cells could impact brain function and may contribute to neurological conditions, including schizophrenia, Tourette’s, and autism.
Constantina Theofanopoulou, a Hunter College researcher who did not take part in the study, says that her research has already benefitted from advances in DNA sequencing. Long reads of vertebrate genomes have helped her uncover the evolutionary history of the receptors and ligands that enable social communication. But long reads obtained from single cells could be a crucial tool to identify genetic variants in the speech disorders that she studies, she says.
However, the technique isn’t perfect, the researchers admit. While better than current methods of scaling up DNA, droplet-based amplification can produce errors and chimeras, where non-neighboring parts of the genome are stuck together. Although they were able to identify and remove chimeras in the current work, their stringent filtering method also threw out correct reads. The team is working on optimizing the conditions to reduce errors during amplification.
“I see this as a proof-of-principle study, where we show that whatever has been done with long-reads in large samples can also be done at the single cell level,” says Ameur. “Now, we can really study single cells in more detail,” he adds.
Like this article? You might also enjoy our monthly Genetics & Genomics newsletter, which is filled with stories like it. You can sign up for free by clicking on the envelope.