New Brain Map Links Mice and Primates to Solve the Rodent-to-Human Research Gap

For decades, a fundamental problem has plagued neuroscience and drug development: a treatment that works brilliantly in a mouse model almost always fails in human trials. The rodent and primate brains diverged roughly 90 million years ago, and that evolutionary gap means many discoveries made in mice simply do not carry over .

Now, scientists at Cold Spring Harbor Laboratory (CSHL) have built the first systematic cross-species brain map designed to bridge exactly that gap. Led by Professor Partha Mitra and colleagues Christopher Mezias, Bingxing Huo, Mihail Bota, and Jaikishan Jayakumar, the team created a quantitative atlas linking mouse brain regions to those of the common marmoset, a New World primate evolutionarily much closer to humans . The full study was published in Communications Biology on August 13, 2026 .

How They Built the Brain Map

Focusing on 'Leaf-Level' Regions

The key insight behind the map is that the most reliable way to compare brains across species is to start at the finest possible level of anatomical detail. The team anchored their comparison on the smallest, most granular subdivisions of the brain — what neuroscientists call leaf-level regions .

While larger brain structures often have inconsistent names and boundaries across different atlases, most existing atlases agree remarkably well on these smallest identifiable chunks. Professor Mitra explained the approach with a simple analogy: "Think of states and cities. New York City might be called New York in one atlas and New Amsterdam in another, but everyone at the local level knows exactly what you're talking about" . By matching these fundamental building blocks first, the team could then reliably build up correspondences between larger brain regions across the two species.

Systematic Literature Synthesis

The team analyzed data from over 100 published studies, each of which used techniques such as neural tracer injections and histological mapping to define brain regions in mice and marmosets . This systematic review allowed them to reconcile inconsistencies between multiple widely used brain atlases for both species.

Quantitative Correspondence Framework

Using computational and statistical pipelines implemented in MATLAB, the researchers compared the spatial, structural, and connectivity profiles of leaf-level regions across mice and marmosets . They established five types of correspondence between brain structures, and found that 43% of mouse leaf-level brain structures have one-to-one correspondences with marmoset structures . The remaining structures are either species-specific or have more complex relationships.

How This Map Solves the Translation Problem

The Evolutionary Gap

The core challenge is that the rodent brain diverged from the primate lineage approximately 90 million years ago . Many drugs and neural circuit findings that show promise in rodents fail in human clinical trials because, at a fundamental level, the brains are wired differently.

Marmosets, as New World primates, share a much more similar brain architecture with humans. By mapping which mouse brain regions have direct homologs in marmosets, the new atlas tells researchers which aspects of rodent-based neuroscience are likely to hold true in primates.

A Practical Translational Bridge

Here is how researchers can use the map in practice:

• If a rodent finding involves a brain region that maps cleanly to a marmoset (and by extension, human) structure, it is a stronger candidate for successful translation .
• If a finding involves a rodent-specific region with no primate counterpart, the finding is unlikely to generalize to humans .

Impact on Drug Development

The map allows researchers to check whether a drug target or neural circuit manipulation that works in a mouse model has an analogous counterpart in a non-human primate brain before moving to expensive human trials. This could help prioritize the most promising drug candidates and reduce the high failure rate of experimental treatments that succeed in rodents but fail in humans .

Broader Context

This cross-species brain map builds on decades of brain-mapping work at CSHL, including the Mouse Brain Architecture Project and contributions to the Brain/MINDS marmoset connectivity atlas in Japan . It provides a unified framework for comparing brain structure across the mammalian lineage.

For researchers developing new treatments for neurological and psychiatric conditions, this map offers a much-needed reality check — and a way to focus resources on the findings most likely to make the leap from mouse to human.

Key Takeaway

43% of mouse leaf-level brain regions have direct one-to-one counterparts in the marmoset brain, according to the CSHL team's analysis . The remaining regions may be species-specific adaptations. This cross-species map gives neuroscientists a systematic tool to decide which rodent findings are worth pursuing in primates — and which ones are likely evolutionary dead ends for human medicine.