Advancing Scientific Knowledge Through Diverse Methods – New Approach Methods (NAMs) and Animal Experiments in Biological and Medical Research
DFG-funded research projects in biology, biomedicine and medicine generally seek to understand how living organisms and their organs function when healthy or diseased. A broad and diverse range of methods and models is available for this purpose. This includes in vitro methods such as cell cultures and organoids, in silico approaches such as computer modelling and AI-based approaches, as well as in vivo models (animal models) and human studies.
Each method and each model has specific potential, but also specific limitations when it comes to addressing specific research questions. Broader insights are often gained by combining findings obtained using different models and methods. Accordingly, a range of complementary methods are generally used in DFG-funded projects.
Animal research and animal models occupy a special position within this spectrum of methods. In the case of many research questions, they are indispensable to gaining a deeper understanding of the complex functioning of organs and organ systems. At the same time, researchers have a particular ethical responsibility towards the animals used. In accordance with the 3Rs principle (replace, reduce, refine), animal experiments may only be used if no equivalent alternative method is available. The number of animals used and the burden placed on them (pain, suffering, distress or lasting harm) must be kept to a minimum.
Promoting the aim to further strengthen animal welfare in research, the funding and further development of what are known as New Approach Methods (NAMs) are becoming increasingly important. While this term is not clearly defined, it is generally understood to encompass a broad range of methods that can reduce, complement or completely replace animal experiments (see (Anchor Link)). NAMs are also becoming increasingly important as a driver of innovation in biological and (bio)medical research. For this reason, targeted funding measures are available at both national and EU level to support method development in this area. By contrast, the DFG’s funding programmes are open to all topics and are geared towards answering specific research questions in the best possible way. The range of methods used is therefore not reported separately. But: How extensively are NAMs used and advanced in DFG-funded projects? And how does their use relate to animal research?
This Data Story provides an insight into the use of NAMs in DFG funding. In order to explore these questions, the Data Story uses a text-mining approach to identify a subset of NAMs in digitally accessible proposal documents based on specific keywords. This includes methods involving organoids and other 3D tissue models, chip technologies and bioprinting. The search terms used in the various keyword categories are listed in (externer Link).
Other methods that fall within the common definition of NAMs (see (Anchor Link)) but cannot be identified using specific keywords are not included (see (Anchor Link)). The Data Story analyses approved proposals under the DFG’s Research Grants Programme and Collaborative Research Centres Programme (including the programme variant Transregio) from 2017 to 2025. These are the most widely used thematically focused funding formats for individual funding (Research Grants Programme) and consortium funding (Collaborative Research Centres Programme), respectively, and are examined here as representative examples.
Use of NAMs under the Research Grants Programme and the Collaborative Research Centres Programme
The text-mining analysis shows that a total of 554 individual proposals under the Research Grants Programme and 163 collaborative proposals under the Collaborative Research Centres Programme involving the use of the NAMs covered by the analysis were approved between 2017 and 2025 (Figure 1). These NAMs are assumed to be used if a proposal contains at least four sentence-level hits from the keyword list (see (Anchor Link)). Collaborative Research Centres (CRCs) are always considered in their entirety; it is not possible to distinguish between individual projects. The vast majority of proposals involving the use of NAMs are attributable to the life sciences (471 Research Grants and 143 Collaborative Research Centres). Smaller numbers are also found in the engineering sciences (63 Research Grants and nine Collaborative Research Centres) and the natural sciences (20 Research Grants and 11 Collaborative Research Centres).
Between 2017 and 2025, the DFG approved 554 individual proposals under the Research Grants Programme and 163 collaborative proposals under the Collaborative Research Centres Programme involving the use of the NAMs covered by this study.
Keyword hits and NAM categories
Analysis of the keyword hits by category (Figure 2) shows that keywords in the organoid category account for the largest proportion of hits by far (69 percent for Research Grants and 75 percent for Collaborative Research Centres). As expected, this proportion is even higher in the life sciences, at 78 percent for Research Grants and 86 percent for Collaborative Research Centres. In the natural sciences and engineering sciences, the categories “Other 3D tissue models”, “Organ-on-chip” and “Bioprinting” account for larger proportions of the hits. This observation suggests that technology development itself is more frequently at the heart of the work programme in these scientific disciplines. These disciplines therefore make important contributions to the development of NAMs that are used primarily in the life sciences.
The proposal-level analysis (Figure 3) shows a clear and steady rise over time in the number of funded research grant projects involving the use of NAMs, from 35 in 2017 to 95 in 2025. At the same time, the number of hits per proposal increases markedly, particularly in the categories of 16–50 and more than 100 sentence-level hits per proposal. This trend can also be observed among Collaborative Research Centres. In 2017, 14 proposals containing four or more sentence-level hits were approved. Although the figures fluctuate from year to year, the overall trend is upwards, reaching 24 approvals in 2025. The number of hits per proposal also rises markedly over time among Collaborative Research Centres, with the increase being particularly pronounced in the category of more than 100 sentence-level hits. This trend suggests that NAMs are being used and further developed with increasing intensity.
Tables 2 and 3 provide an interesting insight into the specific subject areas in which NAMs are used and further developed the most. The tables list the 20 proposals with the highest numbers of sentence-level hits for Research Grants (Table 2) and Collaborative Research Centres (Table 3), respectively. They show that major areas of the life sciences, particularly those of clinical relevance, feature prominently, including cancer, the immune system, the nervous system and metabolism.
The number of approved proposals involving the use of NAMs rose very markedly between 2017 and 2025 (from 35 to 95 Research Grants and from 14 to 24 Collaborative Research Centres). The number of hits per proposal also increased, indicating that NAMs are being used and further developed with increasing intensity. NAMs are used particularly extensively in clinically relevant areas of the life sciences, such as cancer, the immune system, the nervous system and metabolism.
NAMs and animal research in the life sciences
The use and advancement of NAMs is firmly established in DFG-funded life sciences projects. These methods make important contributions, although animal research remains necessary for many research questions in order to reflect the complexity of organs, organ systems and the organism as a whole. Animal research and NAMs are therefore frequently used in a complementary manner in the life sciences. This is illustrated by Figures 4 and 5, which show the use of NAMs in relation to animal research.
Requests for funding under the cost category “Expenses for Laboratory Animals” were analysed as an indicator of animal research. This indicator has certain limitations: funding under this cost category may also be requested for experiments involving invertebrates (e.g. Drosophila, C. elegans), which do not constitute animal experiments requiring authorisation under the German Animal Welfare Act (see (Anchor Link)). In addition, the costs of laboratory animals are sometimes not requested as part of the DFG proposal but are instead covered using the institution’s own budget or by cooperation partners. Nevertheless, analysing this cost category provides an approximate estimate and it is therefore used below as an indicator of animal research.
For Collaborative Research Centres (Figure 4), the analysis shows that NAMs were used in 87 percent of all approved life sciences proposals in 2025 (20 Collaborative Research Centres). In almost all cases (19 Collaborative Research Centres, or 83 percent), funding was also requested under “Expenses for Laboratory Animals”. Only one Collaborative Research Centre involving the use of NAMs (4 percent) did not request funding under “Expenses for Laboratory Animals”. Funding under “Expenses for Laboratory Animals” was requested without NAMs being used in 9 percent of all life sciences proposals (two Collaborative Research Centres). In 2017, this proportion was 52 percent, while the proportion of Collaborative Research Centres using NAMs was 37 percent. Figure 4 reveals a very clear and steady trend over time: the use of NAMs increased, while the number of proposals relying exclusively on animal experiments declined. This trend indicates that NAMs are becoming applicable in a wider range of contexts and that their explanatory power is increasing. The analysis also shows that animal experiments and methods not involving animal experiments are used in a complementary manner in many projects, with due consideration given to their respective potential and limitations.
Compared with Collaborative Research Centres, individual projects under the Research Grants Programme have a narrower thematic focus on specific research questions. Accordingly, the range of methods used within an individual project is also less extensive. Figure 5 shows that throughout the period from 2017 to 2025, the proportion of approved life sciences proposals that neither requested funding under “Expenses for Laboratory Animals” nor used any of the NAMs covered by this study remained relatively constant at around 60 percent. The proportion of projects requesting funding under “Expenses for Laboratory Animals” without using NAMs declined from 38 percent (403 proposals) in 2017 to 29 percent (259 proposals) in 2025. At the same time, the proportion of proposals involving the use of NAMs increased steadily: it was still between 2 and 3 percent from 2017 to 2019, but had risen to 10 percent (86 proposals) of all life sciences proposals decided on in 2025. Half of these projects (5 percent, or 41 proposals) also requested funding under “Expenses for Laboratory Animals”, while the other half (5 percent, or 45 proposals) did not. If proposals involving neither NAMs nor laboratory animals are excluded, the proportion involving the use of NAMs increased from 6.1 percent (26 of 429 proposals) in 2017 to 24.9 percent (86 of 345 proposals) in 2025. A marked increase in the use and further development of NAMs can therefore also be observed under the Research Grants Programme.
NAMs and animal research are frequently used in a complementary manner. NAMs are gaining in importance and explanatory power, but cannot completely replace animal research. NAMs were used in 87 percent of all life sciences Collaborative Research Centres approved in 2025; in almost all of these CRCs, funding was also requested under “Expenses for Laboratory Animals”. Under the Research Grants Programme, approximately 60 percent of life sciences proposals neither include funding under “Expenses for Laboratory Animals” nor involve the use of NAMs. Among the remaining projects, the proportion involving the use of NAMs rose from 6.1 percent in 2017 to 24.9 percent in 2025.
The Data Story provides insights into trends in the use and further development of NAMs in DFG-funded projects since 2017. Given the complementary use of a wide variety of methods, it is neither possible nor objectively warranted to determine and analyse specific funding amounts for individual methodological approaches. Moreover, the analyses presented and the comparison between NAMs and animal research should not obscure the fact that the range of methods used in the life sciences is far more complex and includes numerous other in vitro and in silico approaches, imaging techniques and human studies that are not covered by this analysis.
In summary, the Data Story demonstrates the growing importance of NAMs in biological and (bio)medical research, while also highlighting the importance of a broad and diverse range of methods in which animal research continues to enable insights that cannot be obtained using other methods.
Further Information
The term is not clearly defined and its meaning varies depending on the context. It originated in the field of toxicological and pharmacological testing. Here, NAMs is used as an umbrella term for scientific methods that open up new ways of testing the toxicity or efficacy of materials and substances and investigating the underlying mechanisms. They include the following:
- in vitro methods: e.g. conventional cell cultures, human induced pluripotent stem cells (iPSCs), 3D tissue models, organoids and microphysiological systems (“organ-on-chip systems”)
- ex vivo methods: e.g. studies of tissues taken from an organism, such as human placenta models
- in chemico methods: Chemical test systems that investigate reactions or interactions between substances, for example to analyse molecular reactivity
- in silico methods: Computer-based models, simulations and AI systems
What these approaches have in common is that they can frequently – though not necessarily – avoid or reduce the use of laboratory animals. These approaches frequently use human cells or tissue.
Beyond toxicology, the term NAMs has now become much more widely used across a broad spectrum of research disciplines and fields of application. The research ethics significance of these methods in terms of the 3Rs principle (replace, reduce, refine) is frequently the primary focus. In policy contexts, the abbreviation NAMs is sometimes rendered as “Non-Animal Methods” and used synonymously with the formerly more common term “alternative methods to animal experiments”.
In regulatory contexts involving toxicity and efficacy testing, NAMs are developed and applied in a highly targeted manner. The methods are intended to provide reliable data as a basis for specific decisions, such as assessing the toxicity of substances or the safety of chemicals, medicinal products or consumer products. New methods are therefore frequently developed with a view to complementing or replacing existing testing methods – often animal testing – providing they yield comparable or better information.
In basic research, the development of new methods is strongly driven by the pursuit of knowledge. Many of the methods referred to as NAMs emerge as new experimental tools for investigating biological mechanisms. In some areas of research, for example, organoids are themselves studied in order to investigate genetic and molecular relationships or developmental processes. Basic research is therefore generally geared primarily towards gaining insights into a specific research question. To this end, existing methods are used and refined or new methodological approaches are developed. NAMs are often combined with other methods and used in a complementary manner, including with animal research.
This study covers only a limited selection of methods within the spectrum of NAMs, namely those that can be unambiguously identified using specific keywords. Other methods within the spectrum of NAMs, such as conventional in vitro methods, in silico approaches and imaging techniques, are very widely used in DFG-funded projects but cannot be identified with sufficient precision using keywords and were therefore excluded from the quantitative analysis for methodological reasons.
The following glossary provides more detailed information on methodological aspects of data preparation and analysis.
Search terms and analysis of hits
(externer Link) lists the NAM-specific keywords used to search the proposal documents and indicates how they were assigned to overarching keyword categories.
Certain terms are only counted in combination with contextual keywords occurring in the same sentence (e.g. “spheroid” or “bioprinting” only in combination with terms such as cell culture, tissue, assay or neurons). This approach increases the specificity of the search and reduces false-positive classifications (see (externer Link)).
A sentence-level hit is recorded if a sentence contains at least one valid search term (where applicable in combination with defined contextual terms; see above). The results are aggregated at proposal level. The raw word count or frequency of individual terms within a sentence is not taken into account.
By contrast, the analysis by keyword category in Figure 2 uses keyword hits so as to fully capture sentences containing hits from more than one keyword category: each distinct search term found in a sentence counts as a keyword hit; a single sentence may therefore contain multiple keyword hits.
A minimum threshold of four “sentence-level hits” per proposal was defined for identifying proposals involving the use of NAMs. Hits in bibliographies were removed using a rule-based filter that largely excludes typical reference patterns (including DOI references, parenthesised years, conventional author–year citations and numbered references) so as to prevent distortion caused by mentions occurring solely in citations.
The minimum threshold was determined on the basis of a qualitative review of a sample of proposals with low numbers of hits. This showed that fewer than four sentence-level hits frequently reflected no more than mentions of the keywords, whereas at or above the threshold of four hits, the work programme is highly likely to involve the use of NAMs (generally at project level in the case of Collaborative Research Centres). The text-mining approach does not capture implicit or conceptual references. Nor does it distinguish between the intensity of use, stage of development or degree of innovation of the methods.
Since the search analyses full texts at proposal level, it cannot be ruled out that in individual cases involving an approved Collaborative Research Centre as a whole, individual projects containing sentence-level hits may have been rejected.
Data basis and selection criteria
The analysis covers approved proposals under the Research Grants Programme and the Collaborative Research Centres/Transregio Programme (CRC/TRR) from 2017 to 2025. Other funding programmes and proposals that were rejected or otherwise disposed of are not included. The analysis is based on the full texts of the proposal documents. Joint proposals under the Research Grants Programme are counted only once.
The analysis is based on the full texts of the proposals. An examination conducted as part of the sample analysis showed that proposal abstracts alone frequently do not describe the methods used with sufficient completeness or differentiation and are therefore unsuitable for reliable identification.
DFG subject classification system
The DFG subject classification system distinguishes between a total of four levels: 214 subject areas, 49 review boards, 14 research areas and four scientific disciplines. The report is based on the current subject classification system, valid from 2024 to 2028. The full subject classification system, including the breakdown at the level of all 214 subject areas, is available (Download). The subject classification system maps the operational structures of DFG proposal processing in terms of its subjects and review boards.
Expenses for Laboratory Animals
Requests for funding under the cost category “Expenses for Laboratory Animals” were analysed as an indicator of animal research. This indicator provides an approximation but has certain limitations: it does not capture externally funded expenses for laboratory animals and may also include expenditure relating to invertebrates that do not require authorisation under the German Animal Welfare Act.
Deutsche Forschungsgemeinschaft (2026). Advancing Scientific Knowledge Through Diverse Methods – New Approach Methods (NAMs) and in Biological and Medical Research. DFG Data Stories, Version 1.0, 10.09.2026. (interner Link)
Version 1.0, 10.09.2026
Contact
| E-mail: | Christoph.Limbach@dfg.de |
| Telephone: | +49 (228) 885-2865 |
| E-mail: | richard.heidler@dfg.de |
| Telephone: | +49 (228) 885-2925 |