A team led by Prof. Ali Ertürk, Director of the Institute for Biological Intelligence (iBIO) at Helmholtz Munich and Professor at the LMU, has now developed MouseMapper, a suite of foundation-model-based deep-learning algorithms designed to analyze whole-body biological imaging data. The framework automatically segments 31 organs and tissue types while quantitatively mapping nerves and immune cells throughout the body, enabling comprehensive multi-system analysis in intact mice.

“MouseMapper is built on a foundation model, which means it generalizes far beyond the data it was originally trained on,” says Ying Chen, co-first author of the study. 

Looking Inside an Entire Transparent Mouse

To create whole-body maps, the researchers labeled nerves and immune cells in mice with fluorescent markers visible under the microscope. They then used tissue-clearing techniques to render the animals transparent while preserving the fluorescent signals, allowing imaging deep inside intact bodies.

Using specialized light-sheet microscopy, the team captured detailed three-dimensional images of entire mice, producing datasets containing tens of millions of cellular structures across organs and tissues. MouseMapper then analyzed these data automatically, identifying nerves, immune-cell clusters, and anatomical regions throughout the body.

This allowed the researchers to determine precisely where inflammation and structural damage occur across different tissues – including fat, muscle, liver, and peripheral nerves – without requiring researchers to preselect specific regions of interest.

New Insights Into Obesity, From Mouse to Human

To investigate how obesity reshapes the body, the researchers fed mice a high-fat diet that induced obesity and metabolic dysfunction similar to that observed in humans. Applying MouseMapper revealed widespread changes in both immune-cell organization and nerve architecture across the body. 

One of the most striking findings was a structural change to part of the trigeminal nerve, a major facial nerve that is responsible for facial sensation and motor functions. In obese mice, these sensory nerves had far fewer endings and branches, suggesting a loss of normal nerve function. Behavioral experiments further showed that the animals responded less to sensory stimulation than lean mice, linking the structural damage to impaired sensory function.

The researchers next examined the trigeminal ganglion, the structure containing the cell bodies of facial sensory neurons. Using spatial proteomics, they identified molecular alterations associated with nerve remodeling and inflammation. Remarkably, many of the same molecular signatures were also detected in trigeminal tissue from people with obesity, suggesting that the obesity-associated nerve alterations observed in mice also occur in humans.

“We revealed previously unknown structural and molecular changes in the trigeminal ganglion and its facial branches, and the same molecular signature was conserved in human tissue. This kind of finding simply cannot emerge from studying one organ at a time,” says Dr. Doris Kaltenecker, senior scientist at the Institute for Diabetes and Cancer (IDC) at Helmholtz Munich and first author of the study.

A Platform for Studying Systemic Disease

Beyond obesity, the researchers believe MouseMapper could transform the study of complex diseases that affect multiple organs systems simultaneously, including diabetes, cancer, neurodegeneration and autoimmune disorders. Unlike earlier methods focused on selected organs or tissues, MouseMapper provides an integrated whole-body analysis platform capable of identifying disease “hotspots” throughout the organism.

The team has made whole-body datasets publicly available online, allowing scientists worldwide to explore obesity-associated changes across tissues and organ systems.

“Our goal is to create a comprehensive framework for understanding how diseases affect the body as an interconnected system,” says Ali Ertürk. “Our long-term vision is to build truly realistic digital twins of mice in health and disease: cell-level atlases that we can query, perturb and screen in silico computationally. That would let us pinpoint the earliest changes a disease causes, design interventions to prevent them, and accelerate the discovery of new treatments while reducing the number of physical experiments we need to run.”

Original publication

Kaltenecker et al., 2026: A deep-learning framework reveals whole-body perturbations at cell level. Nature. DOI: 10.1038/s41586-026-10535-2 

Nanocarriers will play a central role in the next wave of life-saving medicines. They enable the targeted delivery of drugs, genes, or proteins to cells within patients. With SCP-Nano, researchers can analyze the distribution of extremely low doses of nanocarriers throughout the entire mouse body, visualizing each cell that has taken them up. SCP-Nano combines optical tissue clearing, light-sheet microscopy imaging, and deep-learning algorithms. First, whole mouse bodies are made transparent. After the three-dimensional imaging of whole mouse bodies, nanocarriers within the transparent tissues can then be identified down to the single-cell level. By integrating AI-based analysis, researchers can quantify which cells and tissues are interacting with the nanocarriers and precisely where this occurs.

Practical Applications of SCP-Nano

Examples of nanocarriers analyzed by Ali Ertürk, the director of the Institute for Intelligent Biotechnologies (iBIO) at Helmholtz Munich, and his team using SCP-Nano include lipid nanoparticles (LNPs), DNA origami structures and adeno-associated viruses (AAVs). These nanocarriers are essential for modern therapeutics that address diseases at their cellular roots, each possessing unique properties that make them suitable for different applications. DNA origami structures are easily programmable and AAVs are highly efficient carriers for gene therapy. The LNPs facilitate RNA delivery, which underlies modern mRNA vaccines and a broad range of other RNA therapeutics. Using SCP-Nano, the researchers demonstrated that DNA origami structures can be preferentially targeted to the immune cells, while AAV variants target distinct brain regions and adipose tissue. Importantly, the platform also revealed that lipid nanoparticles carrying mRNA therapeutics can accumulate in heart tissue. Thus, using SCP-Nano researchers can now detect potentially problematic off-target tissues and associated toxicities before they enter clinical trials, paving the way for the development of safer mRNA therapeutics.

Visualizing Nanocarriers with Unprecedented Precision

“With SCP-Nano, we can detect nanocarriers throughout the body in incredibly low doses, down to 0.0005 mg/kg,” says the study's first author, Dr. Jie Luo. “This gives us an entirely new perspective on how these tiny transport vehicles interact with organs and cells.” Luo emphasizes that it is especially important that SCP-Nano can identify unwanted accumulation in the heart or liver.

The mechanism of nanocarriers is comparable to a parcel delivery service, Ertürk explains: “Each nanocarrier is like a package carrying an important payload that must be delivered to the exact right apartment, not just the one next door. SCP-Nano allows us to track exactly where these packages are delivered, whether they reach their precise intended destination, or if they accidentally end up in unwanted locations.”

Driving Innovation in Personalized Medicine

SCP-Nano allows researchers to precisely identify where nanocarriers accumulate and to visualize their interactions with target cells– a key requirement for safe and effective nanocarrier applications. “SCP-Nano will not only help assess the safety of existing nanocarriers but also drive the development of new, highly targeted applications,” says Luo. “The platform can also help to monitor the success of mRNA therapies or detect potential side effects early.”

A New Era for Drug Development and Personalized Therapies

By combining cutting-edge imaging and AI technologies, SCP-Nano offers researchers and clinicians a new level of understanding of how therapies interact with the body and can easily be extended to human tissues and organs. “Precision medicine and targeted delivery are often discussed, but scalable and effective tools for this have been limited. This new approach offers a solution to a key challenge in drug development,” concludes Prof. Ertürk.

With its potential to minimize side effects and enhance treatment precision, SCP-Nano marks an important step toward safer and more effective therapies in fields like cancer treatment, gene therapy, and vaccine development. This innovation not only addresses major challenges in the development of nanocarrier-based technologies but also drives the future of precision medicine.

A novel AI-powered imaging technique developed by Prof. Ali Ertürk’s team provides new insights into how the SARS-CoV-2 spike protein affects the brain. The method renders organs and tissue samples transparent, enabling the three-dimensional visualization of cellular structures, metabolites, and, in this case, viral proteins. Using this technology, the researchers uncovered previously undetectable distributions of spike protein in tissue samples from COVID-19 patients and mice.

The study, published in the journal Cell Host & Microbe, revealed significantly elevated concentrations of spike protein in the skull’s bone marrow and meninges, even years after infection. The spike protein binds to so-called ACE2 receptors, which are particularly abundant in these regions. “This may make these tissues especially vulnerable to the long-term accumulation of spike protein,” explains Dr. Zhouyi Rong, the study’s first author. Ertürk adds, “Our data also suggest that persistent spike protein at the brain’s borders may contribute to the long-term neurological effects of COVID-19 and Long COVID. This includes accelerated brain aging, potentially leading to a loss of five to ten years of healthy brain function in affected individuals.”

Vaccines Reduce Spike Protein Accumulation and Brain Inflammation

The Ertürk team discovered that the BioNTech/Pfizer mRNA COVID-19 vaccine significantly reduces the accumulation of spike protein in the brain. Other mRNA vaccines or vaccine types, such as vector- or protein-based vaccines, were not investigated. Mice vaccinated with the mRNA vaccine showed lower levels of spike protein in both brain tissue and the skull’s bone marrow compared to unvaccinated mice. However, the reduction was only around 50%, leaving residual spike protein that continues to pose a toxic risk to the brain. “This reduction is an important step,” says Prof. Ertürk. “Our results, while derived from mouse models and only partially transferable to humans, point to the need for additional therapies and interventions to fully address the long-term burdens caused by SARS-CoV-2 infections.” Furthermore, additional studies are needed to evaluate the relevance of these findings for Long COVID patients.

Long COVID: A Societal and Medical Challenge

Globally, 50 to 60 percent of the population has been infected with COVID-19, with five to ten percent experiencing Long COVID. This sums up to approximately 400 million individuals who may carry significant amounts of spike protein. “This is not just an individual health issue – it is a societal challenge,” says Prof. Ertürk. “Our study shows that mRNA vaccines significantly reduce the risk of long-term neurological consequences and offer crucial protection. However, infections can still occur post-vaccination, leading to persistent spike proteins in the body. These can result in chronic brain inflammation and an increased risk of strokes and other brain injuries, which could have substantial implications for global public health and healthcare systems worldwide."

Advances in Diagnosis and Treatment

“Our findings open new possibilities for diagnosing and treating the long-term neurological effects of COVID-19,” says Ertürk. Unlike brain tissue, the skull’s bone marrow and meninges – areas prone to spike protein accumulation – are more accessible for medical examinations. Combined with protein panels – tests designed to detect specific proteins in tissue samples – this could allow for the identification of spike proteins or inflammatory markers in blood plasma or cerebrospinal fluid. “Such markers are critical for the early diagnosis of COVID-19-related neurological complications,” Ertürk explains. “Additionally, characterizing these proteins may support the development of targeted therapies and biomarkers to better treat or even prevent neurological impairments caused by COVID-19.”

Highlighting the broader impact of the study, leading Helmholtz Munich and Technical University of Munich virologist Prof. Ulrike Protzer adds: “Given the ongoing global impact of COVID-19 and the increasing focus on long-term effects, this study, which sheds light on brain invasion pathways and unexpected long-term host involvement, is timely. These critical insights are not only scientifically significant but also of great interest to society.”

Breaking the Wall of Animal Research With AI

Ertürk’s innovative research tackles the limitations of traditional animal experimentation, which, despite being a cornerstone of biological studies, presents significant ethical, logistical, and scientific challenges. By integrating advanced tissue clearing techniques with AI-driven analysis, Ertürk’s team has developed a method to visualize and analyze entire organs and bodies at the cellular level. This approach not only enhances the quality and quantity of data extracted from fewer experiments but also significantly reduces the reliance on animal testing. The societal impact of this research is profound, as it aims to create detailed digital simulations of biological systems, potentially replacing many animal studies altogether, and offering unexpected insights, such as how brain diseases affect the entire nervous system.

“I am deeply honored to receive the 2024 Falling Walls Award in Life Sciences,” states Prof. Ertürk. “This recognition highlights the potential of our work to not only advance scientific understanding but also to address longstanding ethical concerns in biological research. By integrating AI with advanced imaging, we are paving the way for more ethical, efficient, and comprehensive approaches to studying life.”

Many diseases are linked with changes in the expression of certain proteins in the brain. To study those changes, scientists examine how they change during disease progression in model organisms. Imaging entire mouse brains generates vast data sets, necessitating accurate quantification methods for its meaningful interpretation. However, identifying labeled cells within large 3D image data is challenging. While artificial intelligence (AI) holds promise for data analysis, it typically requires extensive data annotation and advanced coding skills, restricting its use to specialized labs. The research team therefore aimed to overcome these barriers, democratizing 3D analysis for broader scientific access.

Virtual Reality Empowering Researchers

To accurately quantify specific cells within brain images, the research team initially trained an AI algorithm to identify them in 3D microscopic images. Leveraging virtual reality (VR) for label generation, the researchers immersed themselves in the images, annotating cells directly in 3D – a faster and more precise method than traditional 2D slice-based approaches. Subsequently, the team employed these VR-generated labels to train an AI algorithm for the automatic identification of active neurons. They integrated the processes of detecting cells, matching them to a brain atlas, and visualizing the results into their DELiVR (Deep Learning and Virtual Reality mesoscale annotation) pipeline. The system operates seamlessly with Fiji, an open-source software for image analysis, in an end-to-end workflow. DELiVR also features a customizable function allowing researchers to train it for specific cell types, such as microglia, an essential immune cell in the brain, demonstrating its adaptability for diverse research projects.

“In essence, DELiVR offers a seamless solution for identifying and analyzing cells throughout the entire brain, providing invaluable insights into their roles and behaviors in both health and disease – all without requiring coding expertise from scientists. DELiVR represents a step towards developing new therapeutic interventions that could ultimately improve the quality of life for individuals affected by debilitating conditions.”

Prof. Ali Ertürk, who led the development of the tool at Helmholtz Munich

Use Case: Cancer-Related Weight Loss

To demonstrate the power of DELiVR, the research team exemplified its capacity to transform our comprehension of how cancer influences our brain activity. Concentrating on the significant clinical challenge of tumor-induced weight loss, they discovered specific brain activity patterns distinguishing cancers that induce weight loss in mice from those that do not. Dr. Doris Kaltenecker, a first author of the study introducing DELiVR, says: “Our findings using DELiVR have revealed potential therapeutic targets within brain regions. This might pave the way for promising strategies to combat cancer-related weight loss in the future.”

About the researchers

Prof. Ali Ertürk, Director of the Institute for Tissue Engineering and Regenerative Medicine at Helmholtz Munich, Professor at the Institute for Stroke and Dementia Research at LMU University Hospital Munich
Dr. Doris Kaltenecker, Postdoc at the Institute for Diabetes and Cancer at Helmholtz Munich, the Translational Diabetes Program at Heidelberg University Hospital, the German Center for Diabetes Research (DZD), and the Institute for Stroke and Dementia Research at LMU University Hospital Munich