SciTransfer
Organization

UNIVERSITEIT LEIDEN

Major Dutch research university combining structural biology, toxicology, social sciences, and growing AI capabilities across 258 H2020 projects.

University research groupmultidisciplinaryNL
H2020 projects
258
As coordinator
133
Total EC funding
€165.2M
Unique partners
1268
What they do

Their core work

Leiden University is a top-tier Dutch research university with exceptionally broad scientific capabilities spanning life sciences, humanities, social sciences, and natural sciences. Their core strengths lie in structural biology (NMR, electron microscopy, X-ray crystallography), toxicology and drug safety, chemical biology, and social science research on migration, democracy, and societal resilience. They run large-scale research infrastructure services for the European scientific community and increasingly apply machine learning and AI to their traditional domains. With over EUR 165 million in H2020 funding across 258 projects, they function as both a fundamental research powerhouse and an infrastructure provider for translational science.

Core expertise

What they specialise in

Structural biology and drug discovery infrastructureprimary
12 projects

Multiple projects including iNEXT (NMR/EM/X-ray infrastructure), high throughput screening projects, and protein crystallography work span the entire H2020 period.

Toxicology, drug safety, and risk assessmentprimary
8 projects

EU-ToxRisk (EUR 5M, coordinated) is a European flagship in mechanism-based toxicity testing; CARDIOTOX and translational safety projects reinforce this cluster.

Migration, democracy, and societal researchprimary
15 projects

Projects like DEMSEC (democratic secrecy), MigrantParents, DISABILITY, and deradicalisation research show sustained social science investment.

Chemical biology and immunologysecondary
8 projects

Crosstag (cross-presentation pathways), IMMUNOSHAPE (carbohydrate immunomodulators), and GLYCOVAX (glycoconjugate vaccines) form a coherent chemical biology cluster.

6 projects

Recent keywords show machine learning, deep learning, and trustworthy AI appearing in the second half of their portfolio — absent from early projects.

Astrochemistry and planetary sciencesecondary
5 projects

EPN2020-RI (Europlanet), EUSPACE-AWE, and multiple astrochemistry-tagged projects show a distinctive space science niche.

Evolution & trajectory

How they've shifted over time

Early focus
Structural biology and metabolomics
Recent focus
AI-augmented life sciences and open science

In the early H2020 period (2015–2018), Leiden focused heavily on structural biology infrastructure (NMR, EM, proteomics), metabolomics, open access infrastructure, and foundational social science research on democracy and adolescence. From 2019 onward, their profile shifted markedly toward computational methods — machine learning, deep learning, network analysis — applied to their existing domains, alongside growing work in co-creation approaches, microbiome research, glycobiology, and open science. The university has maintained its structural biology core throughout but is clearly layering data science and AI capabilities on top of its traditional wet-lab and humanities strengths.

Leiden is rapidly integrating machine learning and AI into its established life science and social science domains, making them an increasingly attractive partner for projects requiring both deep domain expertise and computational sophistication.

Collaboration profile

How they like to work

Role: consortium_leaderReach: Global75 countries collaborated

Leiden coordinates more projects than it joins as a partner (133 coordinated vs 117 as participant), which is unusual for a university and signals strong project leadership capability and administrative capacity. With 1,268 unique consortium partners across 75 countries, they operate as a major European hub rather than relying on a fixed set of repeat collaborators. Their funding scheme mix (heavy ERC, MSCA, and RIA) shows they attract both individual excellence grants and large collaborative research actions — expect a partner that can lead complex consortia but also contribute focused specialist work packages.

Leiden has built one of the largest collaboration networks in H2020, with 1,268 unique partners across 75 countries — covering virtually all EU member states plus significant global reach. This breadth reflects their multidisciplinary nature and makes them a natural connector between otherwise disconnected research communities.

Why partner with them

What sets them apart

Leiden stands out through an unusually balanced combination of hard sciences (structural biology, chemistry, toxicology) and deep humanities/social sciences (migration, democracy, deradicalisation) — most universities of this scale skew heavily toward one or the other. Their transition from traditional lab-based methods to AI-augmented research positions them as a bridge between domain expertise and computational innovation. For consortium builders, their 51% coordination rate and 75-country network mean proven leadership and reach that few European universities can match.

Notable projects

Highlights from their portfolio

  • EU-ToxRisk
    Largest single project (EUR 5M), coordinated by Leiden — a European flagship program for mechanism-based toxicity testing that shaped EU regulatory science.
  • Crosstag
    EUR 1.5M ERC grant in chemical biology of immune cross-presentation, showcasing Leiden's frontier research in immunology at the chemistry-biology interface.
  • DEMSEC
    A philosophical study of secrecy in democratic politics — exemplifies Leiden's distinctive strength in humanities research that few STEM-focused universities pursue at this funding level.
Cross-sector capabilities
Health — toxicology, drug safety, translational medicineDigital — machine learning, AI, high-performance computingSecurity — deradicalisation, societal resilience researchSpace — astrochemistry, planetary science
Analysis note: With 258 projects and rich keyword data across both periods, this is a high-confidence profile. The only caveat is that the 30-project sample skews toward 2015–2016 starts, so the recent-period characterization relies more on keyword shifts than individual project inspection.