As of 2026-08-06: A six-institution US-German consortium has produced the first telomere-to-telomere (T2T), gap-free reference genome of the common marmoset (Callithrix jacchus), a small New World primate that has become one of biomedical research’s most heavily used models for neuroscience and aging. The work, funded by the National Institutes of Health and published in Cell, identifies 76 high-confidence gene candidates linked to Alzheimer’s, Parkinson’s, and related neurodegenerative disease, including previously unseen variation in PSEN1, the gene most strongly associated with early-onset familial Alzheimer’s disease.
A shared resource, not a single-lab finding
The genome was assembled by the UC Santa Cruz Genomics Institute together with the Jackson Laboratory, the University of Pittsburgh, the University of Washington, Oregon Health & Science University (OHSU), the Stowers Institute for Medical Research, and the German Primate Center, as part of the Telomere-to-Telomere (T2T) Consortium’s broader push to build complete reference genomes for the species most relied on in biomedical research. The project was funded by the NIH. Two UC Santa Cruz researchers led the analysis: Prajna Hebbar, a Ph.D. student in the Genomics Institute, and Benedict Paten, professor of biomolecular engineering, working alongside Karen Miga, associate professor of biomolecular engineering and a co-founder of the T2T Consortium.
That six-institution, cross-national structure is the point as much as the genome itself. Rather than a single lab publishing a proprietary dataset, this is explicitly an open, shared community resource: the completed assembly is intended for any researcher working on marmoset biology, aging, or neurodegeneration to use directly, without re-deriving or re-licensing it. For research funders and data managers, it is a working example of the reference-genome-as-shared-infrastructure model the T2T Consortium has already applied to the human and several other genomes — publicly funded, multi-site, and released for community reuse rather than held by any one contributing institution.
What “telomere-to-telomere” actually means
Conventional genome assemblies, including earlier marmoset references, are built by stitching together millions of short sequencing reads — a process that reliably fails in long, highly repetitive stretches of DNA, such as the regions around centromeres and telomeres. Those gaps get left out or papered over with placeholder sequence. A telomere-to-telomere assembly instead uses long-read sequencing technology capable of reading through those repeats directly, producing a genome with no gaps, from one end of each chromosome to the other. The first T2T assembly of a complete human genome was published in 2022; this marmoset genome extends that same gapless standard to a non-human primate model organism for the first time.
The practical payoff is that regions previously invisible to researchers — because short-read assemblies simply could not resolve them — are now available for analysis. That includes the full Major Histocompatibility Complex (MHC), a gene cluster central to immune function that is notoriously repetitive and hard to assemble completely, which the new genome documents in full for the marmoset for the first time.
Why the marmoset, specifically
The common marmoset occupies a specific niche in biomedical research: as a primate, it shares far more genetic and physiological similarity to humans than a mouse does, but at roughly the size of a squirrel it is far cheaper and easier to house and study at scale than a macaque or other Old World primate. Marmosets also naturally show age-related memory decline, which has made them an increasingly common model for studying the biology of aging and neurodegenerative disease specifically — the exact area where this new genome is aimed. A complete, gap-free reference removes a source of systematic blind spots that every marmoset study built on the older, gapped genome has effectively inherited until now.
What the completed genome surfaces for Alzheimer’s and Parkinson’s research
Working from the finished assembly, the consortium identified 76 high-quality gene references tied to Alzheimer’s, Parkinson’s, and related neurodegenerative conditions, and more than 500 genes in the marmoset genome that had not been annotated at all in prior assemblies. Among the specific findings, the team reports novel variation in PSEN1 (presenilin 1), the gene most commonly implicated in early-onset familial Alzheimer’s disease in humans — variation that was not visible in the older, incomplete marmoset reference. According to Hebbar, having a routine, complete T2T-quality genome makes this kind of comparative work more tractable: researchers can now see variation in the same genes implicated in human neurodegenerative disease directly in the marmoset’s own genome, rather than inferring it indirectly. Paten’s team also examined sex-linked genetic differences surfaced by the complete assembly, noting that while such differences may not necessarily translate into functional effects, they are now visible for further study in a way the previous, gapped genome did not allow.
None of this constitutes a treatment or a confirmed disease mechanism on its own. What the complete genome provides is a clearer, gap-free map of candidate genes and regulatory regions for researchers to test experimentally — the kind of foundational resource that downstream functional and comparative studies depend on.
Why this matters for research data management
For research-administration and data-infrastructure audiences specifically, the marmoset T2T genome is a useful case study in how model-organism reference data gets built and shared at scale: a single NIH-funded, multi-institution consortium producing one authoritative, openly reusable dataset rather than each lab maintaining its own partial assembly. That structure reduces duplicated sequencing effort across institutions and gives every subsequent study, in Alzheimer’s and Parkinson’s research or elsewhere, a common, complete reference to align results against — the same rationale that underpins reference-genome infrastructure for humans and other widely studied model species.
The paper is published in Cell (DOI: S0092-8674(26)00815-9). Read the full announcement from UC Santa Cruz.
