Twenty-two years after the completion of the Human Genome Project, scientists have unveiled the most expansive catalog of human genetic variation ever compiled.
Across two new papers published Wednesday (July 23) in the journal Nature, scientists sequenced the DNA of 1,084 people around the world. They leveraged recent technological advancements to analyze long stretches of genetic material from each person, stitched those fragments together and compared the resulting genomes in fine detail.
The results deepen our understanding of “structural variants” within the human genome. Rather than affecting a single “letter” in DNA’s code, such variations affect large chunks of the code — they may be deleted from or added to the genome, or encompass places where the DNA has been flipped around or moved to a different location.
The studies have revealed “hidden” features of the human genome that were previously too technologically challenging to study, said Jan Korbel, the interim head of European Molecular Biology Laboratory (EMBL) Heidelberg, who is a co-author of both new papers. For instance, large portions of the genome contain codes that repeat over and over, and these were thought to be nonfunctional.
“Some 20 years ago, we thought about this as ‘junk DNA’ — we gave it a very bad term,” Korbel told Live Science. “There’s more and more the realization that these sequences are not junk,” and the new work sheds light on these long-maligned DNA sequences.
Additionally, all of the data generated in the new studies are open access, so others in the field can now take “the findings, some of the tools we’ve developed and use them for their purposes to understand the genetic basis of disease,” Korbel told Live Science. “I thoroughly believe that the advances that we’re publishing in Nature today, a subset of these will also make it into diagnostics.”
Related: People’s racial and ethnic identities don’t reflect their genetic ancestry
Over 1,000 genomes
When the first draft of a “complete” human genome was published in 2003, it was actually missing about 15% of its sequence due to technological limitations of the time. In 2013, scientists managed to close that gap by about half. And finally, in 2022, the first “gapless” human genome was published.
In 2023, researchers published the first draft of a human pangenome, which incorporated DNA from 47 people around the world, rather than predominantly being based on one person’s DNA. And that same year, researchers published the first Y chromosome that had ever been sequenced from end to end, because the previous “gapless” genome was still missing the male sex chromosome.
In the past few years, the field has continued to advance, thanks to new technologies and efforts to expand DNA sampling beyond populations of mostly European descent. Those advancements heralded the two papers published in Nature this week.
In the first study, researchers sequenced the DNA of 1,019 people representing 26 populations across five continents. To analyze the DNA, the researchers collected “long reads,” each composed of tens of thousands of base pairs; one base pair corresponds with one rung in the spiral ladder of a DNA molecule.
“With short reads of around 100 base pairs, it is difficult to distinguish between genomic regions that look alike,” explained study co-author Jesus Emiliano Sotelo-Fonseca, a doctoral student at the Centre for Genomic Regulation (CGR) in Barcelona, Spain. That’s especially true in repetitive regions of the genome. “With longer reads, of around 20k base pairs, assigning each read to a unique position in the genome gets much easier,” he told Live Science in an email.
More than half of the new genomic variation uncovered in the study was found in those tricky repetitive regions, including in transposons, also known as jumping genes. Transposons can leap to different locations in the genome, copying and pasting their code. Sometimes, depending on where they land, they can destabilize the genome, introduce harmful mutations and contribute to diseases like cancer.
“Our study reveals that some of these transposons can hijack regulatory sequences to boost their activity, contributing to understanding the biological mechanisms behind their mutagenicity,” or ability to trigger mutations, study co-author Bernardo Rodríguez-Martín, an independent fellow at CGR and a former postdoc in Korbel’s EMBL lab, told Live Science in an email.
The jumping genes can essentially hitch a ride with certain regulatory molecules — long noncoding RNAs — and use that trick to make far more copies of themselves than they usually would. “That’s a very surprising mechanism to us,” Korbel said.
Related: Scientists just discovered a new way cells control their genes
From 95% to 99%
The second study featured far fewer genomes — only 65 in total — but sequenced those genomes more comprehensively than the first study did. The first study captured about 95% of each genome analyzed, while the second study generated 99%-complete genomes.
“It might sound like a small difference, but it’s huge actually from the perspective of the genome scientist,” Korbel said. “To get the last few percentages, it’s a major achievement.”
That leap required different sequencing techniques, as well as new analytical approaches. “This project used cutting-edge software to assemble genomes and identify genetic variation, much of which simply did not exist a few years ago,” co-author Charles Lee, a professor at the Jackson Laboratory for Genomic Medicine, told Live Science in an email.
The sequencing techniques included one that generated long reads with very few errors and one that generated ultralong reads that were slightly more error-prone. At the expense of analyzing fewer genomes, this approach nonetheless enabled the second study to capture stretches of DNA that were totally missed in the first, Rodríguez-Martín said.
Those “hidden” regions included the centromeres, important structures at the centers of chromosomes that are key for cell division. As a cell prepares to split, fibers attach to the centromeres and then pull the chromosome in two. The study found that, in about 7% of centromeres, there are likely two places where these fibers can attach, instead of only one.
“Could that mean that those chromosomes are more unstable? Because if the spindle [fiber] attaches to two points, it might get confused,” Korbel said. That’s a purely speculative idea, he added, but it’s one that can now be explored. The next step will be to study the effects of these centromere variations experimentally, Lee agreed.
Issues with chromosome splitting can lead to various conditions. For example, “Down syndrome is the result of a mistake of chromosome segregation during cell division in meiosis,” when cells split to form sperm and eggs, co-author Dr. Miriam Konkel, an assistant professor at the Clemson University Center for Human Genetics, told Live Science in an email.
Like the first study, the second study also provided an unprecedented look at jumping genes, cataloging more than 12,900. Beyond cancer, jumping genes can also trigger various genetic diseases by causing mutations, as well as prompt more subtle changes in how genes are switched on and off, Konkel noted. A better understanding of the diversity of jumping genes can help unpack their function in human health and disease.
Looking at both studies, scientists can now compare the newly sequenced genomes to other datasets that include both genome and health data, Korbel noted. This would be the first step toward linking the newfound structural variations to tangible health outcomes and, eventually, to incorporating those insights into medical practice.
“Certain clinical studies will not be able to ignore these [sequencing] techniques because they will give them higher sensitivity to identify variation,” Korbel said. “You don’t want to miss variants.”
There’s still more work to be done to improve the genomic data, as well, Lee added. More DNA could be incorporated from underrepresented populations, and the sequencing techniques and software could be further refined to make the process more efficient and accurate. But in the meantime, the pair of new studies marks a major technological feat.
“These advanced tools were developed recently to handle the huge amounts of long-read data we are now using for each genome,” Lee said. “A few years back, assembling a complete human chromosome from end to end, especially including centromeres, was virtually unattainable because the software and algorithms were not mature yet.”