IAP-25-095
Paths to Cooperation: The Combined Effects of Social, Network, and Environmental Factors on the Success of Cooperative Strategies in Bacteria
Cooperation is a fundamental yet fragile feature of many microbial systems. In many bacterial communities, individuals produce “public goods”, such as extracellular compounds that benefit neighboring cells, but these traits are vulnerable to exploitation by non-producers, or “cheaters.” Factors such as the ability to transmit traits horizontally (i.e. horizontal gene transfer), network structure of agents (i.e. number of links, heterogeneity of number of links), and the level of uncertainty (i.e. whether the simulation and environment is allowed to reach equilibrium) been shown individually to affect the development of cooperation, but little is understood about what occurs at intersection of these factors.
Understanding the complex dynamics that allow cooperation to emerge, persist, or collapse remains a central question in evolutionary biology and complex systems research. This project aims to investigate the mechanisms and interactions that drive cooperative behavior in microbial populations, using a combination of laboratory, computational, and statistical approaches.
In one component, laboratory experiments with a model bacterial system will track the dynamics of cooperative and cheating strains over multiple generations, with or without the ability to move the cooperative traits horizontally. These experiments will provide data on how real-world bacteria respond to varying environmental stressors and stability and will inform the next phase.
In another component, multi-agent simulations will be updated iteratively with empirical observations. Each agent will represent a bacterial cell, capable of producing public goods, reproducing, and transferring genes both vertically and horizontally. By systematically varying the ability to horizontally transmit traits, interaction link heterogeneity, and environmental stability, the model will allow exploration of complex interaction effects between these variables that are difficult to control, quantify, and capture experimentally.
Finally, machine learning will be applied to the experimental and simulation data and will extract patterns from complex variable interaction. Unsupervised clustering, supervised classification, and predictive modeling techniques will identify patterns and conditions that consistently promote or inhibit cooperation. This analysis will reveal the key drivers of cooperative persistence, highlighting generalisable principles of population dynamics and emergent behavior.
By combining laboratory experimentation, computational modeling, and data-driven analysis, this project seeks to provide a comprehensive, mechanistic understanding of how cooperative traits arise and stabilize in complex microbial systems. Beyond its immediate biological relevance, the work offers broader insights into emergent phenomena in complex adaptive systems, informing fields from synthetic biology to theoretical ecology.
Research objectives
How does the interaction between horizontal gene transfer and environmental stability affect the development and establishment of cooperation?
How does the interaction of horizontal gene transfer and network structure affect the development of and establishment of cooperation?
How do all three features (i.e. HGT, network structure, and environmental stability) interact to produce cooperative communities?
When including complex interaction effects between features, which variables are the strongest predictors of whether cooperative traits develop and become established in bacterial communities?
What are the most effective pathways towards cooperation, and how do these compare with the reality of the scenarios faced by actual bacteria?
Methodology
The project will involve both laboratory experiments and multi-agent modelling, outlined below:
1. Initially, the student will work in a wet lab to gather data about how real bacteria respond to environmental stressors and changes in networks
2. The output of these laboratory experiments will then be used to inform the creation of artificial bacterial agents with a defined and encoded cooperative trait, as well as the reasonable bounds of variables encoding environmental and network factors
3. A multi-agent model will be developed that incorporates phases of interaction and reproduction and allows for controlled variation in the amount of horizontal gene transfer, network structure, and environmental instability, and where agent success can measurably vary based on interactions with and/or possession of the cooperative trait
4. The student will run simulations in this multi-agent environment, altering the initial conditions and ongoing stability, wherein the incidence, proliferation, and stability of the cooperative trait is tracked over time
5. Finally, clustering techniques will be applied to discover patterns in simulation scenarios that do evolve stable cooperation and feature analysis applied to determine the relative strength of each variable
Project Timeline
Year 1
Literature review
Initial wet lab experiments of cooperative bacteria in non-equilibrium states
Develop initial multi-agent models of bacteria with measurable characteristics for individual, network, and environmental variation
Year 2
Final data gathering from wet lab experiments
Inform multi-agent model with results of wet lab experiments
Code dataset of characteristics and outcomes of bacterial experiments
Year 3
Continuation of multi-agent modeling
Develop dataset that includes variables and outcomes from both laboratory and artificial bacterial experiments
Machine learning-based discovery of variable combinations that lead to established cooperation
Statistical analysis of feature importance
Year 3.5
Final data analysis
Thesis writing and editing
Submission of thesis
Training
& Skills
The student will be embedded in the School of Biology at the University of St. Andrews with significant opportunities to utilize the facilities at the School of Mathematical and Computer Sciences at Heriot-Watt University.
The student will acquire and improve technical skills in computer languages (i.e. Python), modelling techniques, and wet lab skills.
References & further reading
Andras, P., Lazarus, J., & Roberts, G. (2007). Environmental adversity and uncertainty favour cooperation. BMC Evolutionary Biology, 7(1), Article 240. https://doi.org/10.1186/1471-2148-7-240
Dimitriu, T., Lotton, C., Beńard-Capelle, J., Misevic, D., Brown, S. P., Lindner, A. B., & Taddei, F. (2014). Genetic information transfer promotes cooperation in bacteria. Proceedings of the National Academy of Sciences – PNAS, 111 (30), 11103–11108 https://doi.org/10.1073/pnas.1406840111
Iwata, M., & Akiyama, E. (2016). Heterogeneity of link weight and the evolution of cooperation. Physica A, 448(448), 224–234. https://doi.org/10.1016/j.physa.2015.12.047
Lambert, G., Vyawahare, S., & Austin, R. H. (2014). Bacteria and game theory: the rise and fall of cooperation in spatially heterogeneous environments. Interface Focus, 4(4), Article 20140029. https://doi.org/10.1098/rsfs.2014.0029
Lee, I. P. A., Eldakar, O. T., Gogarten, J. P., & Andam, C. P. (2022). Bacterial cooperation through horizontal gene transfer. Trends in Ecology & Evolution (Amsterdam), 37(3), 223–232. https://doi.org/10.1016/j.tree.2021.11.006
Mann, P., Smith, V. A., Mitchell, J. B. O., & Dobson, S. (2021). Random graphs with arbitrary clustering and their applications. Physical Review. E, 103(1), Article 012309. https://doi.org/10.1103/PhysRevE.103.012309
Nagarajan, K., Ni, C., & Lu, T. (2022). Agent-Based Modeling of Microbial Communities. ACS Synthetic Biology, 11(11), 3564–3574. https://doi.org/10.1021/acssynbio.2c00411
Noel, A., Yuting Fang, Nan Yang, Makrakis, D., & Eckford, A. W. (2017). Effect of local population uncertainty on cooperation in bacteria. 2017 IEEE INFORMATION THEORY WORKSHOP (ITW), 2018-, 334–338. https://doi.org/10.1109/ITW.2017.8278046
Strassmann, J. E., & Queller, D. C. (2011). Evolution of cooperation and control of cheating in a social microbe. Proceedings of the National Academy of Sciences – PNAS, 108(Supplement 2), 10855–10862. https://doi.org/10.1073/pnas.1102451108
