Over the course of the project, we established 10 automated ant tracking systems, developed a suite of new methodologies—including dual-colour fluorescent microbead flow cytometry for simultaneous transmission quantification, a general framework for independently manipulating key social network properties, calcofluor staining for in vivo fungal germination timing, and the DAVI pipeline for identity correction in deep-learning pose estimation within small groups of ants.
Using these tools, we carried out a series of large-scale experiments:
• Aim 1: Demonstrated that experimentally altering nest architecture can selectively change network properties, and revealed “architectural immunity”—the first evidence of animals modifying their built environment to reduce future epidemic risk (Science, 2025).
• Aim 2: Showed that organisational immunity is more effective in large colonies than small ones, but only when disease originates from foragers; gene expression analyses confirmed risk-based immune investment.
• Aim 3: Found that ants carrying high pathogen loads avoid high-contact with heterologously contaminated individuals, reducing co-infection risk, and documented pathogen-specific complementarity between individual and social immunity (Nature Communications, 2024).
Additional research uncovered two simple movement rules that explain spatial segregation in social insect colonies (Nature Communications, 2022), identified a pathogen-related vibratory signal (“body shakes”), documented pesticide–pathogen synergies, explored the diffusion of queen pheromones in honeybee colonies (BMC Biology, 2024), and the effect of modular organisation on spatio-temporal collective rhythms.
Results have been widely disseminated via peer-reviewed publications, conference talks, media coverage (including national radio and television), and outreach activities.
