Modeling Engineered Probiotics for Gut Inflammation

ISEF Category: Biomedical Engineering

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

The Hook

What if a probiotic could read gut inflammation and respond on its own? That idea sounds futuristic, but models can test whether it could work before anyone builds it. You can turn a messy disease system into a set of rules, then see how the rules play out. That gives you a real research project without starting in a wet lab.

What Is It?

This project models an engineered probiotic as if it were a tiny decision-maker inside the gut. The bacterium senses inflammatory cytokines, which are immune signals that rise during disease flares, and then releases an anti-TNF mimetic, a molecule designed to block tumor necrosis factor, or TNF. TNF is one of the main signals that drives inflammation in inflammatory bowel disease, so the model asks whether a local response could calm the system down.

Think of the gut like a crowded neighborhood. Cells, microbes, and immune signals all affect one another. A whole-cell agent-based model treats each bacterium or cell like its own actor with simple rules, then watches the whole crowd behavior emerge. BioNetGen and Tellurium help turn those rules into simulations, so you can test how the probiotic might behave across many conditions.

The multi-scale part matters too. You are not just modeling one molecule. You are connecting molecule-level signaling, cell-level response, and population-level gut trends. That is what makes the project feel real instead of purely theoretical.

Why This Is a Good Topic

This is a strong science fair topic because the core question is testable in simulation, but still tied to a real medical problem. You can vary sensing thresholds, release strength, timing, and bacterial survival, then measure how each change affects inflammation outcomes. You also get to compare your model against published gut-microbiome time-series, which gives your project more credibility. A student can learn systems biology, modeling, data validation, and experimental thinking all in one project.

Research Questions

• How does the cytokine-sensing threshold change the timing of anti-TNF release?
• What is the effect of probiotic population size on inflammation suppression over time?
• Does stronger quorum sensing improve or delay the response to an inflammatory flare?
• To what extent does burst release versus steady release change TNF-related signal levels?
• Which model parameter most strongly affects whether the probiotic stabilizes after a flare?
• How does the simulated probiotic response compare with published gut-microbiome time-series patterns?

Basic Materials

• Laptop with enough memory to run simulation software.
• BioNetGen installed through a supported package or local setup.
• Tellurium for building and running dynamic models.
• Python with Jupyter Notebook for analysis and plots.
• Published gut-microbiome or IBD time-series datasets from peer-reviewed papers or public repositories.
• Spreadsheet software for organizing parameter scans and outputs.
• Reference papers on quorum sensing, TNF signaling, and engineered probiotics.

Advanced Materials

• High-performance laptop or workstation for repeated parameter sweeps.
• Access to a university computational biology cluster or server if available.
• Version control system such as Git for tracking model changes.
• COPASI or an equivalent systems biology modeling tool for cross-checking results.
• R or Python libraries for sensitivity analysis and statistical testing.
• Public single-cell, microbiome, or cytokine datasets for validation and comparison.
• Peer-reviewed source papers describing gut inflammation kinetics and synthetic gene circuit behavior.

Software & Tools

• BioNetGen: Builds rule-based biochemical network models for complex signaling systems.
• Tellurium: Runs dynamic systems biology simulations and supports model testing.
• Python: Processes simulation outputs, runs plots, and compares parameter sets.
• Jupyter Notebook: Keeps code, notes, and figures in one place for clear documentation.
• ImageJ: Not used for imaging here, but useful if you later add microscopy-based validation.

Experiment Steps

1. Define the biological question you want the model to answer, such as whether sensing and release can track an inflammatory flare.
2. Choose the smallest set of interacting components that still captures cytokine sensing, bacterial response, and TNF blockade.
3. Translate those components into rules and states in a modeling framework such as BioNetGen or Tellurium.
4. Plan a baseline simulation, then decide which parameters you will vary one at a time and which you will test together.
5. Build a validation plan using published time-series data, and decide which outputs will count as a good match.
6. Prepare a sensitivity analysis so you can identify which design choices matter most in the model.

Common Pitfalls

• Modeling every gut molecule at once, which makes the system impossible to interpret.
• Mixing up TNF concentration with cytokine sensing output, which breaks the logic of the circuit.
• Ignoring bacterial growth and death, which can make the probiotic look unrealistically stable.
• Fitting the model to one paper only, which can hide whether the result generalizes to other datasets.
• Changing multiple parameters at once without a clear plan, which makes it hard to tell which design choice caused the outcome.

What Makes This Competitive

A competitive version of this project would do more than produce a pretty simulation. You would test several circuit designs, compare them with real time-series data, and explain why one design performs better. Strong entries also include sensitivity analysis, uncertainty bounds, and a clear biological reason for every modeling choice. If you can connect model behavior to published gut inflammation patterns, your project starts to feel like real research instead of a class demo.

Project Variations

• Model the same circuit in Crohn’s disease data instead of a broad IBD dataset.
• Swap the anti-TNF mimetic for a different anti-inflammatory output and compare the predicted dynamics.
• Add microbiome competition terms and test how native gut bacteria change probiotic stability.

Learn More

• NCBI PubMed: Search for review articles on TNF signaling, inflammatory bowel disease, and engineered probiotics.
• NIH NCBI Bookshelf: Read free textbook chapters on immune signaling and systems biology methods.
• BioNetGen documentation: Learn rule-based modeling concepts and how to build reaction networks.
• Tellurium documentation: Find examples for dynamical systems biology simulation and model comparison.
• NCBI GEO: Search for public gene expression and time-series datasets related to gut inflammation and microbiome change.
• MIT OpenCourseWare: Look for free systems biology or computational biology course materials that explain modeling basics.