E. Coli GyrA Fitness Landscapes Under Antibiotics

ISEF Category: Microbiology

Subcategory: Microbial Genetics  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

A single DNA letter change can flip a bacterium from vulnerable to hard to kill. That tiny swap can also open, or block, the path to stronger resistance. You can model that path without growing dangerous bacteria. This project turns evolution into something you can map and test.

What Is It?

This project studies how one small change in a gene can change antibiotic resistance. The gene, gyrA, helps make a bacterial enzyme that fluoroquinolone antibiotics attack. A single nucleotide polymorphism, or SNP, means one DNA base changes to another. That can alter one amino acid in the protein, which can change how well the drug binds.

Think of the protein like a lock and the antibiotic like a key. A mutation can slightly bend the lock, so the key does not fit as well. But some mutations also hurt the bacteria itself. That tradeoff matters. A mutation that blocks the drug but breaks the enzyme too much may become a dead-end. A different mutation may help the bacterium survive just enough to reach a better mutation later. That kind of path is called a stepping-stone.

Why This Is a Good Topic

This is a strong science fair topic because you can ask a clear, testable question about real antibiotic resistance, and you can answer it with models and public data instead of a wet lab full of risky cultures. You can compare mutation effects, simulate selection under different drug levels, and look for patterns that explain which mutations spread. The project connects to medicine, public health, and evolution. You can also build real skills in structural biology, coding, and data analysis.

Research Questions

• How does the Ser83 mutation change predicted gyrA stability under different amino acid substitutions?
• What is the effect of fluoroquinolone selection pressure on the fixation probability of single gyrA mutants?
• Does a double mutant act as a stepping-stone or a dead-end across a range of antibiotic gradients?
• To what extent do Rosetta ΔΔG scores predict which gyrA variants have higher simulated fitness?
• Which mutation order gives the highest probability of resistance evolution in a Wright-Fisher model?
• How does population size change the chance that a weakly beneficial gyrA mutation survives drift?

Basic Materials

• Laptop or desktop computer with enough RAM for Python analysis.
• Python installed with NumPy, pandas, SciPy, and matplotlib.
• Rosetta access through an academic installation or a school account.
• PubMed and NCBI Gene access for background literature and sequence data.
• Reference sequence for E. coli gyrA from NCBI.
• Spreadsheet software for tracking simulation outputs and summary tables.
• External storage or cloud backup for large output files.

Advanced Materials

• University workstation or cluster access for repeated Rosetta runs.
• Rosetta software suite for protein modeling and ΔΔG estimation.
• Python environment with Biopython, NumPy, pandas, SciPy, matplotlib, and statsmodels.
• Sequence alignment tools for comparing gyrA variants.
• Structural visualization software such as PyMOL or ChimeraX.
• Access to NCBI, PubMed, and any institutional sequence databases.
• Version control system such as Git for code tracking.
• Statistical software for model comparison and sensitivity analysis.

Software & Tools

• Python: Runs the Wright-Fisher simulation, processes outputs, and makes plots.
• Rosetta: Estimates how each mutation changes protein stability and binding-related features.
• PyMOL: Helps you inspect where Ser83 sits in the protein structure.
• NCBI Gene: Provides the reference gyrA sequence and annotation.
• PubMed: Helps you find review articles and primary papers on fluoroquinolone resistance.

Experiment Steps

1. Define the exact mutation set you will compare, including single and double gyrA variants around Ser83.
2. Gather the reference sequence, protein structure, and any known resistance data to anchor your model.
3. Build a scoring plan that turns each mutation into a fitness estimate from stability and drug pressure.
4. Set up a Wright-Fisher simulation that tracks how mutation, selection, and drift change variant frequencies over generations.
5. Test several antibiotic gradient scenarios so you can see when a mutation becomes a stepping-stone versus a dead-end.
6. Compare simulated outcomes with published resistance patterns and adjust your assumptions if the model conflicts with reality.

Common Pitfalls

• Treating Rosetta ΔΔG as a direct measure of resistance, which confuses protein stability with drug survival.
• Ignoring epistasis, which means the effect of one mutation changes when another mutation is present.
• Using only one antibiotic level, which hides how selection changes across a gradient.
• Forgetting population size, which can make a beneficial mutant disappear by chance in the simulation.
• Comparing model results to too few published variants, which makes your conclusions look stronger than the evidence supports.

What Makes This Competitive

A class-level project often stops at one mutation and one simple graph. A stronger project compares many mutational paths, checks how sensitive the answer is to selection strength, and tests whether the same ranking holds across different population sizes. You can also separate structural effects from evolutionary effects, which adds depth. A careful validation against published resistance mutations makes the work feel much more like real research.

Project Variations

• Focus on other quinolone target genes, such as parC, and compare whether their mutation paths differ from gyrA.
• Replace the binary resistant versus sensitive score with a continuous growth fitness metric to model weaker selection.
• Add codon-level mutation bias to see whether some resistance routes appear more often because they are easier to generate, not just fitter.

Learn More

• NCBI Gene: Search for gyrA in Escherichia coli to find the reference sequence, gene annotation, and linked records.
• PubMed: Search for review articles on fluoroquinolone resistance, gyrA, and topoisomerase mutations.
• NCBI Protein: Find the gyrA protein record and compare annotated domains across bacterial species.
• Rosetta documentation: Read the official guides for ΔΔG prediction and protein mutation modeling.
• MIT OpenCourseWare: Search for population genetics and evolutionary dynamics lectures that explain selection, drift, and fixation.