Predict CRISPR Off-Target Effects in Tomato

ISEF Category: Plant Sciences

Subcategory: Genetics and Breeding  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

One wrong DNA match can send CRISPR to the wrong spot. That matters even in a plant like tomato, where a small edit can change growth, color, or flavor. You can study that risk without a wet lab. Your project can ask which published guides look safest before anyone edits a plant.

What Is It?

CRISPR uses a guide RNA, which is a short sequence that leads the Cas enzyme to a target DNA site. Think of it like a GPS pin for the genome. If the guide matches the wrong place too well, CRISPR can cut there too. That wrong cut is called an off-target effect.

In this project, you do not edit tomatoes yourself. You use published CRISPR guides for the tomato SlMYB gene and run them through a free prediction tool, CRISPOR. The tool compares each guide to the tomato genome and estimates where extra cuts might happen. You can then compare guides by predicted specificity, mismatch pattern, and off-target score. That gives you a real bioinformatics project with a clear question and measurable outputs.

Why This Is a Good Topic

This is a strong science fair topic because the core question is testable with public data and free software. You can compare multiple published guides, rank them, and see whether some design choices lead to fewer predicted off-target sites. That connects to a real problem in crop genetics, since breeders and plant scientists want edits that stay precise. You can learn genome browsing, guide scoring, data organization, and basic statistical comparison without needing a lab bench.

Research Questions

• How does guide RNA length affect predicted off-target count in tomato SlMYB targets?
• What is the effect of GC content on CRISPOR specificity scores for published tomato guides?
• Does the location of a guide within the SlMYB gene change its predicted off-target profile?
• To what extent do mismatch positions near the PAM affect off-target risk in tomato genome predictions?
• Which published SlMYB guide has the highest predicted specificity across the Solanum genome?
• How does the number of predicted off-target sites differ between guides that target coding regions and guides that target regulatory regions?

Basic Materials

• Laptop or desktop computer with internet access.
• Google Sheets or Excel for data tables.
• Free CRISPOR web access or API access.
• PubMed or Google Scholar access for finding published SlMYB guides.
• Tomato genome reference links from a public database such as NCBI or Ensembl Plants.
• Notebook for tracking guide IDs, scores, and filters.

Advanced Materials

• Laptop or desktop computer with internet access.
• Python installed locally or in a notebook environment.
• Pandas for data cleaning and summary tables.
• Biopython for sequence handling and basic parsing.
• R or Python plotting tools for guide comparison charts.
• Genome browser access for Solanum lycopersicum reference annotations.
• Command-line tools for batch API queries and result storage.

Software & Tools

• CRISPOR: Predicts CRISPR off-target sites and scores guide specificity for your tomato sequences.
• NCBI Genome Data Viewer: Helps you confirm where each guide sits in the tomato genome.
• PubMed: Lets you find published SlMYB CRISPR guide sequences and supporting papers.
• Google Sheets: Organizes guide scores, off-target counts, and comparison tables.
• Python: Automates data cleaning, scoring summaries, and plots if you want a larger dataset.

Experiment Steps

1. Define the exact comparison you want to make, such as guide specificity, off-target count, or mismatch pattern.
2. Collect published SlMYB guide sequences from papers and record each guide with its source and target region.
3. Match each guide to the tomato genome reference and verify that you are using the same genome version across all runs.
4. Run each guide through CRISPOR and store the predicted off-target sites, specificity scores, and related metrics.
5. Choose the summary statistics and visualizations that answer your question, such as ranking, clustering, or score comparisons.
6. Plan a validation check by comparing your results against the guide design choices reported in the original papers.

Common Pitfalls

• Mixing genome versions, which changes predicted off-target sites and makes guides hard to compare.
• Using guide sequences that include extra cloning bases, which can break the CRISPOR input and distort scores.
• Comparing guides from different papers without tracking the same target type, which creates a mismatch in your analysis.
• Treating predicted off-targets as proven edits, which overstates what the computer output can actually tell you.
• Skipping a clean data table, which makes it easy to lose sequence names, source papers, and score values.

What Makes This Competitive

A stronger project goes beyond one guide and one score. You can compare several published guides, test whether the tool agrees with the authors' design logic, and look for patterns in where risky off-targets appear. You can also add a tougher analysis, like grouping off-target sites by mismatch position or by genomic region. That turns a simple search task into a real study of guide design quality.

Project Variations

• Compare SlMYB guides against guides for another tomato gene to see whether off-target risk differs by target gene.
• Analyze whether guides with similar GC content still get very different specificity scores.
• Test whether guides aimed at coding regions have different predicted off-target patterns than guides aimed at promoter regions.

Learn More

• CRISPOR documentation: Search for the CRISPOR guide design and off-target prediction documentation to learn how its scoring works.
• NCBI Genome Data Viewer: Use the public genome browser to inspect tomato gene locations and annotations.
• Ensembl Plants: Search the tomato genome page to confirm gene models, transcripts, and genomic context.
• PubMed: Search for review articles on CRISPR specificity, guide design, and off-target prediction in plants.
• MIT OpenCourseWare: Search for free genomics or bioinformatics course materials that explain sequence alignment and genome analysis.