Autoimmune Enhancer Variant Detector

ISEF Category: Biomedical and Health Sciences

Subcategory: Genetics and Molecular Biology of Disease  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: Full Year

The Hook

Most disease-linked DNA changes do not sit inside genes. They often land in the switches that control genes, and those switches can act only in certain tissues. You can use public ChIP-seq and GWAS data to hunt for autoimmune variants that may hijack those switches. That turns a pile of anonymous SNPs into a testable research question.

What Is It?

Think of a gene as a lamp and an enhancer as the wall switch that turns it on or off. An enhancer is a stretch of non-coding DNA that helps control when a gene gets used. In autoimmune disease, a variant in that switch can shift immune activity without changing the protein-coding part of the gene itself.

ChIP-seq is a way to map where proteins or chromatin marks sit on DNA, so you can spot active regulatory regions in a chosen tissue. GWAS, or genome-wide association studies, point to DNA variants linked to a trait or disease. ABC-score, short for activity-by-contact score, tries to predict which enhancer talks to which gene, so you can rank variants by how likely they are to matter in a specific tissue.

Why This Is a Good Topic

This topic works well for a science fair because you can test clear, measurable ideas with public data. You can compare disease variants against tissue-specific enhancer maps, then ask whether autoimmune hits cluster in the immune tissues that should matter most. You will learn how to clean genomic data, set controls, and judge whether a pattern is real instead of random.

Research Questions

• How does tissue type change the number of autoimmune GWAS variants that overlap active enhancers?
• What is the effect of using different ChIP-seq histone marks on the number of candidate causal variants?
• Does ranking variants by ABC score improve overlap with known autoimmune loci?
• To what extent do autoimmune-associated variants cluster in B-cell, T-cell, or blood enhancers?
• Which autoimmune diseases show the strongest enrichment of non-coding variants in tissue-specific enhancers?
• How does nearest-gene assignment compare with enhancer-gene pairing for prioritizing candidate variants?

Basic Materials

• Computer with internet access and at least 8 GB of RAM.
• Spreadsheet software such as Google Sheets or Excel.
• Python or R with plotting packages.
• Free access to the UCSC Genome Browser, GWAS Catalog, and ENCODE Portal in your browser.
• PubMed for reading review articles on enhancers and disease variants.

Advanced Materials

• Linux workstation with 16 to 32 GB of RAM.
• Command-line tools such as bedtools, liftOver, and tabix.
• Jupyter Notebook or RStudio for analysis and plotting.
• Local reference genome build files and gene annotation files.
• Processed ChIP-seq peak files or signal tracks from ENCODE.
• A tissue-specific enhancer atlas and a curated autoimmune variant list.

Software & Tools

• UCSC Genome Browser: Lets you inspect variants, genes, and regulatory tracks in one view.
• GWAS Catalog: Helps you find published autoimmune associations and study metadata.
• ENCODE Portal: Provides tissue-specific ChIP-seq and chromatin data for enhancer mapping.
• Python: Cleans variant tables, joins annotation files, and runs ranking scripts.
• R: Makes enrichment plots, correlation tests, and summary charts.

Experiment Steps

1. Define one autoimmune disease set and one tissue set, then decide how you will map variants to enhancers.
2. Build a consistent pipeline for pulling public GWAS hits, enhancer annotations, and ABC scores.
3. Set rules for what counts as a candidate causal variant, including overlap, linkage, and tissue match.
4. Choose a null model or control set so you can test whether your hits are enriched beyond random variants.
5. Plan how you will compare nearest-gene assignment with enhancer-gene pairing and score-based ranking.
6. Decide how you will summarize uncertainty, multiple testing, and top candidates for follow-up.

Common Pitfalls

• Mixing genome builds between GWAS and ChIP-seq files, which makes valid overlaps disappear.
• Treating every non-coding SNP in linkage disequilibrium as causal, which inflates your hit list.
• Using enhancer tracks from the wrong tissue, which turns tissue-specific signals into noise.
• Ranking only by ABC score, which can hide variants with strong chromatin evidence but weaker model output.
• Forgetting to match ancestry or study background in controls, which can make enrichment tests look stronger than they are.

What Makes This Competitive

A stronger version does more than list overlapping SNPs. It compares several autoimmune diseases, checks whether the same enhancer logic repeats across tissues, and tests your ranking against a matched control set. You can also compare nearest-gene calls with enhancer-gene calls, then measure how often they disagree. Careful controls and cleaner statistics make the project feel like real research.

Project Variations

• Focus on one autoimmune disease, such as type 1 diabetes, and compare enhancer hits across pancreatic and immune tissues.
• Swap ABC scores for another enhancer-gene method, then see whether the same variants stay near the top.
• Restrict the search to variants in high-linkage disequilibrium blocks and test whether the candidate list gets cleaner or noisier.

Learn More

• ENCODE Portal: Search tissue-specific ChIP-seq, histone mark, and accessibility data for enhancer maps.
• GWAS Catalog: Find autoimmune associations, SNP lists, and study metadata for your disease set.
• UCSC Genome Browser: Compare variants, genes, and regulatory tracks in one browser view.
• NIH PubMed: Search review articles on enhancers, fine mapping, and non-coding disease risk.
• NIH Roadmap Epigenomics Mapping Consortium: Explore reference chromatin state maps for immune and other tissues.
• MIT OpenCourseWare: Look for free genomics and bioinformatics lectures when you need a refresher on sequence annotation and statistics.