Originally published 2022-04-03.

On 31 March 2022, an international group of researchers called the Telomere-to-Telomere (T2T) consortium published the first truly complete, gap-free sequence of a human genome in the journal Science. It filled in the roughly 8% of human DNA that the famous Human Genome Project had left unread when it was declared finished in 2003.
The new sequence, known as T2T-CHM13, spans about 3.055 billion DNA letters and covers every chromosome from one end, or telomere, to the other, apart from the Y chromosome, which followed in 2023. Here is what was missing, how the team filled it, and what the result does and does not mean.
Why the 2003 genome was not truly complete
The Human Genome Project was a landmark, but its reference sequence focused on the gene-rich, easier-to-read parts of our DNA. Some regions were simply too repetitive for the technology of the time.
Older sequencing machines read DNA in short fragments, a few hundred letters at a time, which computers then stitch together like a jigsaw. That works well when each piece is distinctive. It fails in stretches where the same pattern repeats thousands of times, because short pieces from different copies look identical and cannot be placed. The result was a reference with gaps, mostly around the centromeres, near the telomeres, and on the short arms of five chromosomes.
Successive updates narrowed those gaps, but the standard reference used by labs and hospitals, known as GRCh38, still had about 8% of the genome unresolved.
How the T2T consortium closed the gaps
The breakthrough came from long-read sequencing, which reads much longer stretches of DNA in one go. The team combined two approaches:
- Oxford Nanopore reads, which can run to around a million letters at once with modest accuracy, ideal for spanning huge repeats.
- PacBio HiFi reads, about 20,000 letters long but almost perfectly accurate, ideal for getting each letter right.
They also chose their source DNA cleverly. Instead of an ordinary sample, which carries two slightly different copies of each chromosome, one from each parent, they used a cell line called CHM13 that carries essentially one version of each chromosome. With no second copy to tell apart, the assembly became far simpler.
The consortium was led by Karen Miga of the University of California, Santa Cruz, Evan Eichler of the University of Washington and Adam Phillippy of the US National Human Genome Research Institute (NHGRI), whose group led the final finishing work. The results appeared as six main papers in Science, with companion studies in other journals.
What was found in the missing 8%
The newly added DNA amounts to nearly 200 million letters, roughly the size of a whole chromosome. Among the highlights:
- Centromeres in full. These are the constricted regions where a chromosome's two copies are held together and where the cell's machinery attaches to pull them apart during cell division. They are made of vast repeated arrays that had never been fully read.
- New genes. Reports described around 2,000 new gene predictions. Most are thought to be inactive copies, but around 100 may code for proteins.
- Better variant detection. Using the complete sequence as a reference, researchers found more than 2 million additional genetic variants in other people's genomes, and more accurate variant information in 622 genes already known to be medically relevant.
- Corrected errors. The new reference fixed mistakes in older versions and produced better maps of five chromosome arms that had been especially hard to read.
NHGRI director Eric Green called it "an incredible scientific achievement" that gives the first comprehensive view of our DNA blueprint.
Common misunderstandings
Headlines about "the complete human genome" can give the wrong impression. A few points worth keeping straight:
- It is one genome, not everyone's. T2T-CHM13 is a reference, a single high-quality map. Every person differs from it in millions of small ways, and one sequence cannot capture human diversity.
- The 2022 version had no Y chromosome. Because the CHM13 cells carry an X but no Y, the Y chromosome had to be completed separately. The T2T team published a complete Y in 2023, adding more than 30 million letters missing from the older reference.
- It is not a cure. A complete map makes difficult regions available for research, but understanding what each part does and turning that into treatments takes years.
- Centromeres are about cell division, not directly about ageing. Some early coverage linked them to longevity. Their known role is in separating chromosomes correctly when cells divide; errors there are linked to certain diseases, but anti-ageing treatments are not an expected near-term result.
What came next: the pangenome
To address the "one genome" problem, a related effort, the Human Pangenome Reference Consortium, released a draft pangenome in 2023 built from 47 genomes of people from varied ancestries. The idea is to replace a single reference with a collection of complete or near-complete genomes that better reflects how human DNA varies around the world.
Together, the complete T2T sequence and the pangenome mark a shift from asking "what does a human genome look like?" to asking "how do human genomes differ, and why does it matter?"
Why it matters for medicine and research
Many genes involved in immune responses and in how people respond to drugs lie in or near regions that were previously missing or wrong. A more accurate reference means fewer errors when labs compare a patient's DNA with the standard, and fewer variants missed because they fell into a gap.
Adam Phillippy put it this way: "Now that we can clearly see everything, we are one step closer to understanding what it all means." He compared finishing the sequence to putting on a new pair of glasses.
For individuals, the change is gradual rather than dramatic. Clinical genetic testing relies on established references and validated methods, and it takes time for new references to be adopted. If you are thinking about genetic testing for a health reason, talk to a doctor or a genetic counsellor, who can explain what a test can and cannot tell you.
Sources: NHGRI: Researchers generate the first complete, gapless sequence of a human genome (31 March 2022); Nurk et al., The complete sequence of a human genome, Science 376 (2022); Wikipedia: Human genome (Y chromosome and pangenome).

COMMENTS