Mice dominate laboratory research worldwide, making up roughly 60% of all animals used in scientific studies across Europe, with rats adding another 14% on top of that. But the full roster of research animals spans an enormous range, from transparent fish embryos and fruit flies to pigs, ferrets, and a small number of non-human primates. Each species fills a specific scientific niche, chosen not arbitrarily but because its biology offers something no other species, and no current technology, can replicate for a particular question.
Mice and Rats Run the Show
In the European Union’s most comprehensive count, mice accounted for about 61% of the roughly 11.5 million animals used in research, while rats made up just under 14%.1PMC Central. Principles of Animal Research for Graduate and Undergraduate Students The numbers in the United States follow a broadly similar pattern, though exact comparisons are harder because U.S. reporting excludes purpose-bred mice and rats from its official count under the Animal Welfare Act. Still, every estimate puts rodents firmly at the top.
Why mice? Three reasons converge. First, their genomes are well-mapped, and researchers can manipulate specific genes with precision, “knocking out” a gene entirely or switching it on only in a particular tissue or at a particular point in development to see what happens.2PubMed Central. The Mighty Mouse: The Impact of Rodents on Advances in Biomedical Research Second, mice breed quickly and are small enough to house in large numbers, making studies with hundreds of animals logistically feasible. Third, most human disease genes have counterparts in rodent genomes. Genes tied to neurological function show especially strong conservation between species, which is why rodent models of brain diseases tend to mirror human disease processes more closely than models of some other conditions.3PubMed Central. Evolutionary conservation and selection of human disease gene orthologs in the rat and mouse genomes
Rats fill a complementary role. They are large enough for surgical procedures that would be impractical in mice, and researchers have historically preferred them for behavioral and toxicology studies because their responses to drugs and stressors track more predictably. The same gene-editing tools now available in mice have increasingly been applied to rats as well, which has expanded their utility in recent years.2PubMed Central. The Mighty Mouse: The Impact of Rodents on Advances in Biomedical Research
Zebrafish and the Power of Transparency
Zebrafish have become one of the fastest-growing model organisms in biomedical science, and the reason is almost literal: you can see through them. Their embryos develop outside the mother, grow rapidly, and are naturally transparent, which lets researchers watch organs form, blood flow, and tumors develop under a microscope in real time. This makes them uniquely suited for drug screening, because you can expose hundreds of embryos to candidate compounds and observe the effects at the whole-organism level without needing to dissect anything.4PubMed Central. A versatile, automated and high-throughput drug screening platform for zebrafish embryos
As zebrafish grow, they normally develop pigment that blocks visibility. Researchers have gotten around this in two ways: by using genetically mutant strains that stay transparent into adulthood, or by chemically treating embryos to suppress pigment formation. The fully transparent “Casper” line, for example, was created by combining two pigment-related mutations, eliminating both the reflective cells in the skin and the pigment-producing cells entirely.5FEMS Microbiology Reviews. Diving into drug-screening: zebrafish embryos as an in vivo platform for antimicrobial drug discovery and assessment This has been a game-changer for infection research, because scientists can inject bacteria into the fish and then track the infection’s spread through the body over several days, watching how the immune system responds and how antibiotic candidates perform in a living organism.
Zebrafish share about 70% of their genes with humans, which is more overlap than most people expect. Their hearts, kidneys, and immune systems share enough structural similarity with ours to make them useful models for studying everything from heart disease to antimicrobial resistance. And because a single breeding pair can produce hundreds of embryos at a time, the throughput is orders of magnitude higher than what is possible with mammals.
Fruit Flies, Worms, and Other Invertebrates
Not all research animals have backbones, and some of the most powerful insights in biology have come from organisms most people would never associate with medical research. The fruit fly, Drosophila melanogaster, has been a workhorse of genetics research for over a century. Its genome is compact, its life cycle is about two weeks, and roughly 75% of known human disease genes have recognizable counterparts in the fly. Scientists studying everything from cancer to sleep to aging routinely use fruit flies because genetic experiments that would take years in mammals can be completed in weeks.
Recent work has continued to reveal surprising complexity in fly biology. A 2025 study found that when fruit flies were given the choice to move between dark and illuminated areas in a constant-light environment, they could maintain or even recover rhythmic behavioral patterns, and these self-imposed rhythms were accompanied by molecular changes in the clock neurons that drive circadian behavior.6PubMed. Fruit flies actively restart their circadian clock by proactively shaping their environment Findings like this are relevant to human medicine because the same core clock genes operate across species, so understanding how circadian rhythms break down in flies can inform research on sleep disorders and the health effects of shift work in people.
The roundworm C. elegans occupies an even more stripped-down niche. An adult has exactly 959 cells, and every one of them has been mapped. Researchers know the fate of every cell during development, which makes the worm ideal for studying how cells decide to live, divide, or die. The core molecular pathway that controls programmed cell death was first worked out in C. elegans, including the roles of specific proteins that have direct equivalents in human cells.7ScienceDirect. Methods for detection and analysis of apoptosis signaling in the C. elegans germline That discovery earned a Nobel Prize and has shaped our understanding of how cancer cells resist dying.
Non-Human Primates Fill a Narrow but Critical Gap
Monkeys make up about half of one percent of all animals in research, but their role is disproportionately important in areas where no other species can substitute. The main reason is immunological: primate immune systems are close enough to ours that they can reproduce the full biological process of a human infection, from initial exposure through immune response and either recovery or disease progression. This makes them essential for vaccine development, because a vaccine that works in mice may fail in humans if the immune response is fundamentally different, and primates catch that gap.8PubMed Central. The Critical Role of Nonhuman Primates in Medical Research
Rhesus macaques, the most commonly used primate species, have been central to research on HIV/AIDS, Ebola, Zika, and more recently mpox. In a 2025 study, for instance, researchers tested a new mpox vaccine candidate in rhesus macaques and found it induced stronger neutralizing antibodies than the existing licensed vaccine, and the antibodies provided better protection when transferred to mice.9PubMed Central. Mpox multiprotein virus-like nanoparticle vaccine induces neutralizing and protective antibodies in mice and non-human primates This kind of head-to-head comparison between a new candidate and an approved product often relies on primate models because the immune response in rodents would not capture the same dynamics.
Primates are also uniquely valuable in neuroscience. Their brains share structural and functional features with ours that no rodent brain can replicate, which matters for research on Alzheimer’s, Parkinson’s, and depression. The ethical bar for using primates is higher than for any other species, and approval processes are lengthy and heavily scrutinized, but the scientific community generally considers them irreplaceable for a specific subset of questions where the biology of other animals falls short.8PubMed Central. The Critical Role of Nonhuman Primates in Medical Research
Pigs, Dogs, and Ferrets
Large animal models each serve a distinct purpose that smaller animals cannot. Pigs are anatomically similar to humans in the size and structure of their organs, their physiology, their immune systems, and even their genomes. That combination makes them valuable for surgical research, for testing medical devices like heart valves or stents at human scale, and increasingly as potential sources of organs for xenotransplantation, where a pig organ is transplanted into a human.10PubMed Central. Importance of the pig as a human biomedical model Pig skin is also a surprisingly close match for human skin, which makes pigs useful in wound-healing and burn research.
Dogs occupy a different niche, primarily in pharmaceutical safety testing. Regulatory guidelines in many countries require that new drugs be tested in both a rodent and a non-rodent species before human trials can begin, and dogs have historically been one of the standard non-rodent choices. Industry groups have collected case studies illustrating instances where toxicity signals appeared in dogs but not in rodents, catching safety problems before human volunteers were exposed.11PubMed. The Dog as a Second Species for Toxicology Testing Provides Value to Drug Development The use of dogs in research is one of the most publicly contentious aspects of animal testing, and alternative second species like minipigs are increasingly being explored where the biology supports it.
Ferrets fill a surprisingly specific gap in infectious disease research. Their respiratory tracts respond to influenza viruses in ways that closely mirror human infection, including how the virus transmits through the air. Researchers use ferrets to assess how likely a new flu strain is to cause a pandemic, by testing whether the virus can spread between animals housed in the same room but without direct contact. Studies have shown that ferrets infected with highly transmissible flu strains exhale more virus-containing aerosol particles than ferrets infected with strains that do not spread easily, providing a way to rank the pandemic potential of emerging viruses.12PubMed Central. Comparison of the levels of infectious virus in respirable aerosols exhaled by ferrets infected with influenza viruses exhibiting diverse transmissibility phenotypes This model is extensively used by public health agencies worldwide as an early-warning system.13PubMed Central. Ferrets as Models for Influenza Virus Transmission Studies and Pandemic Risk Assessments
Frogs and the Study of Early Development
Amphibians, particularly the African clawed frog Xenopus laevis and the Mexican axolotl, have long been staples of developmental biology. Their eggs are large compared to mammalian cells and develop externally, making it possible to observe the earliest stages of embryonic development directly. After fertilization, the eggs undergo a rapid series of synchronous cell divisions before the embryo’s own genome switches on and development becomes more complex.14ScienceDirect. Cell cycles during early steps of amphibian embryogenesis: A review This window of synchronized division has been critical for understanding how cell cycles work and how errors in early development lead to birth defects.
The axolotl is also famous for its regenerative abilities, being one of the few vertebrates that can regrow entire limbs, parts of its heart, and even sections of its spinal cord. Researchers studying regeneration hope that understanding the molecular signals the axolotl uses might eventually point toward therapies that could improve healing in humans, who have largely lost this capacity.
The Ethics Framework That Governs Animal Research
Virtually every country that conducts animal research operates under some version of the “3Rs” framework: Replacement (using non-animal methods when possible), Reduction (using the fewest animals necessary to get reliable results), and Refinement (minimizing pain and distress). The 3Rs are embedded in regulations from the European Union’s Directive 2010/63 to the U.S. Animal Welfare Act and its implementing guidelines. In practice, researchers must justify their use of animals through an ethics review process before experiments begin.
How well this works in practice, though, is a fair question. An analysis of ethics applications submitted in Sweden found that the information researchers provided about harm, benefit, and their consideration of the 3Rs was often insufficient, and sometimes missing entirely, making it difficult for review committees to properly weigh the costs against the expected benefits.15PubMed Central. Approved Ambiguities: An Analysis of Applications for the Ethical Review of Animal Research in Sweden-Focusing on Harm, Benefit, and the 3Rs A comparative legal analysis of Finland and Switzerland reached a similar conclusion from a different angle: while both countries incorporate the 3Rs extensively into their regulations, the framework mostly governs what happens during experiments and only partially addresses issues like the breeding of surplus animals or what happens to animals after a study ends. The authors argued that without stronger institutional accountability and a clearer legal path toward non-animal methods, the 3Rs risk functioning mainly as a tool for making existing animal research more humane rather than as a genuine mechanism for moving beyond it.16LEOH – Journal of Animal Law, Ethics and One Health. Cross-Jurisdictional Study of the 3Rs: A Comparative Legal Analysis of Animal Research Governance in Finland and Switzerland
Housing Conditions Affect the Science, Not Just the Animals
One thing that often surprises people outside the field is that how you house a lab animal can change your experimental results. Environmental enrichment, giving animals things like nesting material, shelters, running wheels, or objects to manipulate, is now recommended or required by most institutional guidelines. This is partly an animal welfare measure, but it also has scientific consequences. Studies on rats have shown that enrichment can produce measurable changes in brain chemistry and behavior, which means that the “standard” barren cage that was once the default may have been introducing an unrecognized variable into decades of neuroscience research.17PubMed. The impact of environmental enrichment in laboratory rats–behavioural and neurochemical aspects
Enrichment programs need to be designed carefully, though. Providing an enrichment that is too variable between facilities can actually make results harder to reproduce. The goal is to give animals a more natural experience while keeping conditions standardized enough that a lab in Tokyo and a lab in Boston can compare their results. Guidance from the field emphasizes that enrichment must be evaluated both for its benefit to the animal and its impact on experimental outcomes.18PubMed. Environmental enrichment for laboratory rodents and rabbits: requirements of rodents, rabbits, and research Even the gut microbiome of lab rodents, which varies depending on the vendor, the facility, and what the animals eat, has been identified as a major source of irreproducibility between studies.19PubMed Central. Microbiota and reproducibility of rodent models
Organ-on-a-Chip and the Push Toward Alternatives
The most significant recent shift in the landscape of animal research is not about a new animal model. It is about whether certain experiments need animals at all. Organ-on-a-chip technology uses tiny microfluidic devices lined with living human cells to mimic the function of a specific organ, a lung, a liver, a gut, or even a miniature “body-on-a-chip” that links several organs together. These chips can simulate drug metabolism, immune responses, and disease processes using human tissue rather than animal tissue, which in principle gives more relevant data for predicting how a drug will behave in people.20PubMed Central. Organ-on-a-Chip: A New Paradigm for Drug Development
The technology is advancing quickly but has real limitations. Current chips can replicate some key features of human tissues, including physical microenvironments and cell-to-cell interactions, but they cannot yet capture the full complexity of a living organ, let alone a whole organism.21Fundamental Research. Advances in human organs-on-chips and applications for drug screening and personalized medicine In neuroscience, for instance, researchers have developed 3D brain organoids and microfluidic systems that model aspects of neurodegenerative diseases, but no current system can reproduce all the key features of conditions like Alzheimer’s or Parkinson’s.22PubMed. Beyond animal models: revolutionizing neurodegenerative disease modeling using 3D in vitro organoids, microfluidic chips, and bioprinting The chips are best understood as a supplement to animal research rather than a full replacement, at least for now.
A major regulatory milestone came in late 2022, when the U.S. FDA Modernization Act 2.0 was signed into law. The original 1938 law had mandated animal testing for every new drug entering development. The 2022 update removed that blanket requirement, allowing drug developers to use alternatives like organ-on-a-chip systems, computer modeling, or other non-animal methods if they can demonstrate the data is sufficiently predictive.23PubMed. FDA Modernization Act 2.0 allows for alternatives to animal testing The law does not ban animal testing. It simply removes the assumption that animals are the only valid path forward, opening the door for newer technologies to prove themselves.
Why Not Just Use Computers?
A question that comes up frequently, especially in public discussions about animal welfare, is why researchers cannot simply simulate everything on a computer. The honest answer is that biological systems are too complex for any current model to simulate reliably at the whole-organism level. A computer can predict how a drug molecule will bind to a single protein, and it can flag obvious toxicity risks based on chemical structure. But predicting what that molecule will do once it enters a living body, gets metabolized by the liver, crosses the blood-brain barrier, interacts with gut bacteria, and triggers immune responses in multiple tissues simultaneously, is beyond what any computational model can do with the accuracy needed to make life-or-death decisions about human safety.
Machine learning is closing some of these gaps. Deep-learning approaches have been used, for example, to evaluate how well genetic sequences are functionally conserved between humans and pigs, revealing that both rapidly evolving and slowly evolving DNA sequences showed higher functional similarity between the two species than sequences evolving at a neutral rate.24Cell Genomics. Deep learning-based evaluation of genomic conservation at the functional level between humans and pigs This kind of work helps scientists choose the right animal model for the right question, potentially reducing the number of animals used by steering researchers toward the species most likely to give relevant results. But identifying which animal to use is very different from eliminating the need for animals altogether. For the foreseeable future, the direction of travel is toward using fewer animals, using them more wisely, and replacing them where the technology allows, rather than toward a world where animal research has disappeared entirely.