In the last three years, the number of human genomes that have been sequenced (their DNA read letter by letter) has jumped from a handful to hundreds, with thousands more in progress. But all of those genome readings lack some crucial information. A person inherits two copies of each chromosome, one maternally and one paternally. Existing sequencing methods do not indicate whether genetic variations that lie close to each other on the genomic map were inherited from the same parent, and therefore come from the same chromosome, or if some lie on the maternal chromosome and some on the paternal one. Knowing this has a variety of uses, from sequencing fetal DNA to more easily detecting the genes responsible for different diseases to better tracking human evolution.
Now two teams have devised ways to determine these groupings—known as the haplotype—in an individual. Stephen Quake and collaborators at Stanford University developed a way to physically separate the chromosome pairs and sequence each strand of DNA individually. Jay Shendure and colleagues at the University of Washington in Seattle sequenced DNA from single chromosomes in specially selected pools and used this information to piece together the genome. Both projects were published this week in Nature Biotechnology.
“It was a real technical flaw in the genomes [sequences] that have been published to date,” says Quake, a bioengineer at Stanford who was one of Technology Review’s top innovators under 35 in 2002. “Every genome we are going to do from now on going will be recorded with the haplotype.”
Quake’s team capitalized on microfluidics technology that they have developed for separating and analyzing single cells. First, the researchers trapped single cells during a specific phase of the cell cycle in which the two copies of its chromosomes are split apart. Then they burst open the cell, randomly partitioned chromosomes into different chambers on a microfluidics chip, and copied, or amplified, and analyzed the DNA in each chamber.
Shendure, a TR35 winner in 2006, and his team amplified 40,000 letter stretches of DNA randomly sampled from individual chromosomes. Because each piece of DNA comes from one half of a chromosome pair, researchers know that all the genetic variants within its sequence lie on the same chromosome.
Shendure and Quake say that having haplotype information will have an enormous impact on human genetics, helping not only to diagnose and understand the genetic basis of some diseases but also to track the evolution of our species from primate ancestors.
If someone has two disease-linked mutations within a single gene, it’s difficult to determine with current genome sequencing methods if there is one genetic mistake on the maternal copy and one on the paternal copy or if both variations lie within the same copy of the gene. In the former case, the person has two defective genes, which are likely to cause health problems. In the latter, the person has one good copy of the gene and one bad copy. In many cases, having the good copy can compensate for the defective one.
Haplotyping also makes it possible to determine a person’s human leukocyte antigen (HLA) type, from immune genes that must be closely matched between donor and recipient in cases of bone marrow or organ transplant. “It’s one of the most polymorphic [variable] parts of human genome,” says Quake. Current methods to determine HLA type generate a list of variations but give no information about which of them lie on which chromosome. “If you don’t keep track of this, you may not able to get a perfect match,” says Quake. “We showed you can measure [the haplotype] and get information that in principle can be used for better matching for bone marrow transplants.”
The technology might also be used to sequence fetal genomes from DNA collected from the mother’s blood, in order to detect genetic abnormalities. (The DNA in the fetal blood is a mix of the mother’s and the child’s, making it particularly difficult to generate a whole genome sequence.)
Beyond medicine, researchers say, haplotype information will aid research in population genetics, such as estimating the size and timing of human expansions and migrations. “You can capture diversity to higher resolution if you have individual chromosomes,” says Nicholas Schork, a geneticist at the Scripps Research Institute who was not involved in either project and wrote a commentary on the research for Nature Biotechnology. “You lose a lot of information if you look at things at a genotype level versus a haplotype level.”
Researchers have been able to statistically infer haplotype for European populations, thanks to the fact that Europeans went through a genetic bottleneck thousands of years ago. (Haplotypes very gradually grow shorter, as the chromosome pairs break and reassemble with each generation. Europeans have long haplotypes that haven’t yet broken down, making them easier to analyze.) But statistical techniques have not worked for African populations, meaning that genetic information for this group is much sparser. For this reason, most of the genome-wide association studies done to date have focused on European populations.
Both approaches add to the cost of genome sequencing, so it’s not clear how quickly they will catch on, Schork says. “Shendure’s approach is one people could likely implement in labs now,” he says. Quake’s approach generates much more complete data—a haplotype that is the length of an entire chromosome—but it is technically more challenging, requiring specialized chips to analyze the single cells. “Single cell sequencing and the ability to separate chromosomes in a dish is complicated,” says Schork. “Unless someone builds an affordable assay, it won’t be used routinely.” Quake says that the chips that his lab and close collaborators use are currently being built at an academic foundry at Stanford. He says, “Perhaps there will be a commercial solution at some point.”
How Facebook and Google fund global misinformation
The tech giants are paying millions of dollars to the operators of clickbait pages, bankrolling the deterioration of information ecosystems around the world.
DeepMind says it will release the structure of every protein known to science
The company has already used its protein-folding AI, AlphaFold, to generate structures for the human proteome, as well as yeast, fruit flies, mice, and more.
Inside the machine that saved Moore’s Law
The Dutch firm ASML spent $9 billion and 17 years developing a way to keep making denser computer chips.
This is what happens when you see the face of someone you love
The moment we recognize someone, a lot happens all at once. We aren’t aware of any of it.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.