Kombucha Cellulose Wound Dressing Design

ISEF Category: Biomedical Engineering

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

The Hook

A wound dressing can do more than cover skin. It can help fight infection, hold moisture, and act like a tiny engineered scaffold. Kombucha bacteria already make cellulose, so students can treat a familiar kitchen culture as a starting point for biomaterial design. The hard part is choosing the right antimicrobial peptide and checking whether it fits the job.

What Is It?

This project asks you to think like a biomaterials designer. Kombucha SCOBYs contain bacteria that build cellulose, which is a natural fiber. In research, cellulose can serve as a dressing material because it is flexible, water-friendly, and easy to shape into a thin film or pad. You can picture it like a paper towel that is being redesigned for skin contact instead of cleanup.

The synthetic biology part adds a second layer. An antimicrobial peptide is a short protein-like molecule that can damage or stop bacteria. Your job is not to build a medical product at home. Your job is to design and compare peptide candidates on the computer, then predict which ones might work best when paired with a cellulose-based dressing. APD3 is a peptide prediction resource, and AlphaFold-Multimer helps estimate how proteins may interact in complexes.

Why This Is a Good Topic

This is a strong science fair topic because you can turn a big medical problem, wound infection, into a clear design question. You can compare peptide candidates, rank them with computational tools, and justify your choices with data. The project also connects to real needs in wound care, infection control, and biomaterials. A student can realistically learn bioinformatics, protein structure thinking, and experimental design without needing to build a full wet-lab product.

Research Questions

• How does peptide length affect APD3 classifier scores for antimicrobial activity?
• What is the effect of net positive charge on predicted antimicrobial ranking?
• Does hydrophobicity change the interaction score between candidate peptides and cellulose-related binding targets?
• To what extent do AlphaFold-Multimer interaction scores agree with APD3 rankings for the same peptide set?
• Which peptide features best predict a strong antimicrobial score without raising predicted aggregation risk?
• How does the choice of peptide family change the top-ranked candidates for a wound dressing design?

Basic Materials

• Laptop or desktop computer with internet access.
• Spreadsheet software such as Google Sheets or Excel.
• PubChem or UniProt database access for peptide and protein reference data.
• APD3 peptide database or similar antimicrobial peptide prediction resource.
• AlphaFold-Multimer results access through a public server or published structures, if available.
• Notebook for recording candidate sequences, scores, and decision rules.

Advanced Materials

• Workstation or cloud access for sequence analysis and structure modeling.
• Python with Biopython, pandas, and matplotlib.
• Jupyter Notebook for reproducible analysis.
• Access to AlphaFold-Multimer or a comparable protein complex modeling pipeline.
• Access to BLAST or HMMER for sequence similarity checks.
• Molecular visualization software such as PyMOL or UCSF ChimeraX.

Software & Tools

• Python: Organizes peptide data, calculates summary features, and compares candidate rankings.
• Jupyter Notebook: Keeps code, notes, and plots in one place for a reproducible workflow.
• ImageJ: Measures any future gel or colony images if you later pair the design study with lab validation.
• PyMOL: Visualizes peptide structure and predicted binding regions.
• Google Sheets: Tracks candidate peptides, scoring outputs, and design decisions.

Experiment Steps

1. Define the design goal by deciding what makes a peptide a good fit for a cellulose wound dressing.
2. Build a candidate library from known antimicrobial peptides, then remove sequences that are too similar or too risky for your comparison set.
3. Choose the features you will compare, such as length, charge, hydrophobicity, and predicted structure.
4. Rank the candidates with APD3-style classification outputs, then compare those ranks with AlphaFold-Multimer interaction results.
5. Plan a scoring rubric that balances antimicrobial potential, structural confidence, and likely manufacturability.
6. Select a final shortlist and explain why each candidate fits the dressing concept better than the others.

Common Pitfalls

• Treating APD3 scores as proof of real-world antimicrobial activity, which they are not.
• Comparing peptides with very different lengths, which can make the ranking look unfair.
• Ignoring peptide solubility or aggregation risk, which can make a strong-looking candidate unrealistic.
• Using AlphaFold-Multimer outputs without checking whether the binding partner choice matches the dressing design question.
• Mixing sequence sources with unclear annotation, which can introduce duplicates or mislabeled peptides.

What Makes This Competitive

A stronger project does more than rank a few peptides. It builds a clear scoring system, compares multiple design criteria, and explains why one candidate wins over another. You can raise the level by testing whether APD3 and structure-based scores agree, then checking where they disagree. A competitive project also names its limits and shows how the design could move toward lab validation later.

Project Variations

• Compare peptides for different wound types, such as chronic wounds, burns, or surgical dressings.
• Swap the cellulose scaffold for another biomaterial, such as chitosan, alginate, or gelatin, and compare the design tradeoffs.
• Focus on a resistance-aware analysis by screening candidate peptides for features linked to lower bacterial escape risk.

Learn More

• PubMed: Search review articles on antimicrobial peptides, wound dressings, and cellulose biomaterials.
• NIH NCBI Bookshelf: Find free background chapters on protein structure, host defense peptides, and biomaterials.
• UniProt: Look up peptide and protein annotations, sequences, and functional notes.
• PubChem: Check molecular properties and related compound data for biomaterial and peptide references.
• MIT OpenCourseWare: Search for free molecular biology, bioengineering, and computational biology course materials.