Phage-Host Coevolution Models

ISEF Category: Computational Biology and Bioinformatics

Subcategory: Computational Evolutionary Biology  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

Viruses and bacteria can lock each other in an endless arms race. One side gains an edge, the other side evolves back. Your simulation can test when that cycle turns into kill-the-winner dynamics, and when it looks more like a Red Queen race.

What Is It?

This project models how bacteriophages, which are viruses that infect bacteria, and their hosts change over time in response to each other. Think of it like two teams playing tag, where every rule change by one side forces the other side to adapt. In some runs, the fastest growing bacterial type gets hit hard, which fits the idea of kill-the-winner. In other runs, both sides keep changing in step, which fits the Red Queen idea.

You can build this with Avida, a digital evolution platform, or with Python if you want full control. A phylodynamic model tracks how populations and their genetic lineages change together through time. That lets you ask not just who wins, but how the pattern of change shifts when you change mutation rate, infection strength, or population structure.

Why This Is a Good Topic

This is a strong science fair topic because you can test clear variables, generate real data from simulations, and compare two famous evolutionary patterns. It connects to viral evolution, antibiotic resistance, and host-pathogen dynamics, so the work feels current and useful. You can learn simulation design, data analysis, and how to turn a biological idea into a computational model.

Research Questions

• How does mutation rate change whether kill-the-winner or Red Queen dynamics dominate?
• What is the effect of host population size on oscillation strength in phage-host coevolution?
• Does higher infection efficiency increase the frequency of winner crashes in the host population?
• To what extent does spatial structure change the balance between short boom-bust cycles and long coevolutionary cycles?
• Which parameter settings produce the most stable coexistence between phage and host lineages?
• How does the presence of multiple host resistance traits change lineage turnover over time?

Basic Materials

• Laptop or desktop computer with Python installed.
• Python packages for simulation and plotting, such as NumPy, pandas, and Matplotlib.
• Spreadsheet software for organizing output data.
• Avida digital evolution software, if you choose that platform.
• Text editor or IDE, such as VS Code or Spyder.
• External drive or cloud storage for backing up simulation runs.

Advanced Materials

• University or lab workstation with higher memory and faster processing.
• Python scientific stack, including SciPy, statsmodels, and seaborn.
• Avida source files or a custom agent-based simulation codebase.
• R for time-series analysis and model comparison.
• Git for version control.
• Access to journal articles and review papers on phage-host coevolution.

Software & Tools

• Python: Runs custom simulation code and handles data analysis.
• Avida: Simulates digital evolution and lets you test coevolution rules.
• Jupyter Notebook: Keeps code, figures, and notes in one place.
• R: Supports time-series testing and statistical comparison of dynamics.
• ImageJ: Measures plot features if you export figures for visual comparison.

Experiment Steps

1. Define the biological question your model will answer, then choose whether you will use Avida or a custom Python framework.
2. Select the few parameters you will vary, and keep the rest fixed so you can interpret cause and effect.
3. Build rules for host growth, phage infection, resistance, and mutation, then check that the model behaves sensibly in a baseline run.
4. Plan outputs that capture both population size and lineage change, so you can compare kill-the-winner and Red Queen patterns.
5. Design a control set that isolates one factor at a time, then add paired runs for each scenario you want to compare.
6. Decide how you will classify each run, using plots, summary statistics, and a prewritten rule for which dynamic dominates.

Common Pitfalls

• Using too many free parameters, which makes it impossible to tell which one caused the pattern you see.
• Tracking only population size, which misses lineage turnover and hides the Red Queen signal.
• Building infection rules that are too simple, which can create fake oscillations that do not reflect coevolution.
• Comparing runs with different random seeds but no replicate plan, which makes chance look like biology.
• Changing model structure halfway through the project, which breaks your ability to compare results across conditions.

What Makes This Competitive

A strong version of this project goes beyond making pretty oscillation plots. You would compare multiple model assumptions, use enough replicates to test whether the pattern holds, and define a clear rule for classifying each run. You could also add a novel twist, such as spatial structure, multiple resistance genes, or a sensitivity analysis that shows which parameters matter most. That turns the project from a demo into a real model test.

Project Variations

• Use a lattice-based spatial model instead of a well-mixed population to test whether local contact changes the dominant coevolution pattern.
• Compare a custom Python simulation with Avida outputs to see whether both frameworks predict the same host-phage dynamics.
• Add a second host resistance trait and test whether trait diversity delays winner crashes or strengthens Red Queen cycling.

Learn More

• NCBI Bookshelf: Search for free chapters and review-style textbooks on coevolution, population genetics, and host-pathogen dynamics.
• PubMed: Search review articles on bacteriophage-host coevolution and phylodynamics.
• NIH NCBI Taxonomy and Genome resources: Use these databases to understand phage and bacterial lineages.
• MIT OpenCourseWare: Look for free materials on evolutionary biology, computational modeling, and probabilistic systems.
• Avida-ED: Find the educational version of Avida and its documentation through the official project site.
• Philosophical Transactions of the Royal Society B: Search for review papers on kill-the-winner dynamics, Red Queen theory, and phage ecology.