🌟 Fun & Detailed Theoretical Summary of the Lecture (Endometriosis Genetics + GWAS)

1. 🌺 What is Endometriosis?

Endometriosis is a chronic condition where tissue resembling the endometrium (inner lining of the uterus) grows outside the uterus. These growths:

• Contain glands + stroma, similar to normal endometrium
• Can grow on ovaries, fallopian tubes, peritoneum, and even near the bowel
• Establish their own blood and nerve supply
• Bleed internally, causing inflammation, adhesions, and pain

Symptoms

• Chronic pelvic pain
• Fatigue
• Infertility
• GI disturbances (mimics IBS)
• Headaches, depression

Why diagnosing is hard

• Symptoms vary widely
• Overlap with multiple conditions
• Gold-standard diagnosis is laparoscopy
• Average delay to diagnosis: 6–10 years

Epidemiology & Impact

• Affects 1 in 9 women (Australia estimate)
• Strong hereditary component: first-degree relatives → 2–3× increased risk
• Huge economic burden: $7–9 billion AUD annually
• No known cure → management focuses on surgery, hormone therapy, lifestyle support

2. 🧬 Why Study Genetics in Endometriosis?

About 50% of the risk appears due to genetic factors, based on twin studies.

Understanding genetics helps researchers:

• Identify biological pathways behind the disease
• Find causal genes
• Develop better therapies
• Improve diagnostic predictions
• Separate disease into meaningful subtypes

3. 🧪 From Old Genetics to Modern GWAS

🔙 Early days (1980s–1990s)

Researchers used restriction fragment length polymorphisms (RFLPs) to map big-effect mutations in monogenic diseases (e.g., cystic fibrosis). But endometriosis isn't monogenic — it involves many small-effect variants.

🚀 Enter GWAS (Genome-Wide Association Studies)

Key innovation: Use genotyping chips with ~800,000 SNPs spread across the entire genome.

How a GWAS works

1. Genotype many individuals (cases + controls)
2. For each SNP, test: “Is allele A vs. B more common in cases?”
3. Adjust for massive multiple testing → significance threshold = 5×10⁻⁸
4. Plot results in a Manhattan plot 🏙️
   • Peaks = regions with significant association
   • Nearby SNPs cluster due to linkage disequilibrium

Why LD matters

A significant SNP is usually not the causal variant — it’s a “tag” for a region. Causal fine-mapping is needed to find the true functional variant.

4. 🧬 Twin Studies: Proof of Heritability

Two key twin studies (1999 and later replication) showed:

• Identical (MZ) twins share 100% of their DNA
• Fraternal (DZ) twins share ~50%

Higher concordance in MZ → ~51% heritability.

This finding justified building large international genetic consortia.

5. 🌍 International Endometriosis Genetics Consortium

Growth of sample size over the years led to dramatic increases in discoveries.

Most recent study:

• 42 genomic regions identified
• 49 independent signals
• 31 new since previous publications
• Replicated past hits → confirms robustness of GWAS design

6. 🔍 The Big Challenge: From Variant → Gene → Biology

Most GWAS variants lie in non-coding regions:

• Intergenic
• Intronic
• Regulatory DNA (enhancers, promoters)

These influences typically modify gene expression, not protein sequence.

This creates two major challenges:

1. Which variant is causal? Many linked SNPs cluster around a region.
2. Which gene is affected? Often multiple nearby genes could plausibly be impacted.

7. 🎛️ Gene Regulation Matters

Gene expression is controlled by:

• Enhancers
• Promoters
• Insulators
• DNA methylation
• Chromatin structure

Projects like ENCODE and NIH Roadmap map these, but tissue coverage is incomplete.

Endometrium adds complexity

• Gene expression changes dramatically across the menstrual cycle
• Case–control differences may be subtle
• Disease biology may be cell-type-specific, especially stromal vs epithelial cells

8. 🧫 eQTLs and the Search for Regulatory Effects

An expression quantitative trait locus (eQTL) is a genetic variant affecting gene expression.

Researchers combine:

• GWAS results (variants affecting disease risk)
• eQTL results (variants affecting expression)

To test whether the same variant explains both.

Three possible models

1. Variant A affects expression; Variant B affects disease (independent)
2. One variant affects both, but expression isn’t causal
3. The shared variant changes gene expression, which causes disease ← desired model

Statistical colocalization helps choose between these models.

9. 🧩 Case Examples from the Paper

🧬 Example 1: Chromosome 15 (SRP14 region)

• SNP associated with endometriosis
• Also alters gene expression in endometrium and blood
• Suggests a potential causal regulatory mechanism
• Variant may influence methylation too

🧬 Example 2: Chromosome 12 (FGD6 / VST)

A SNP lies in a bi-directional promoter, altering both neighboring genes. Both genes have plausible biological roles in endometriosis.

10. 🧬 Splicing QTLs (sQTLs)

Not only expression changes matter—variants can alter:

• Splice sites
• Transcript isoforms
• Exon usage

Many disease associations may actually be mediated through splicing, not expression.

11. 🧬 Why Haven’t We Found All Causal Genes Yet?

Key limitations:

1. Insufficient sample sizes

• eQTL studies often have 200–1000 samples
• Only detect large effects, many of which are non-functional
• True disease mechanisms likely involve small, tissue-specific effects

2. Tissue and cell-type specificity

GWAS variants may only act in:

• Certain menstrual-cycle phases
• Stromal cells
• Epithelial cells
• Immune cells in lesions

Bulk tissue hides these nuances.

3. Evolutionary constraints

• True disease-causing variants tend to be under purifying selection
• Large eQTLs usually are not — meaning they probably aren’t biologically critical
• So overlap between GWAS hits and eQTL hits is smaller than expected

12. 🧬 Biological Pathways Emerging from the Genes

When combining all likely causal genes, patterns appear:

• Reproductive system development
• Cell proliferation
• Epithelial cell differentiation
• Regulatory pathways known to malfunction in endometriosis lesions
• Genes acting in multiple pathways and multiple cell types

Small genetic effects can accumulate → meaningful biological impact.

13. 🌐 Overlap With Other Conditions

Endometriosis shares genetic components with:

• Pain disorders
• GI diseases
• Depression
• Ovarian cancer

This has clinical implications:

• Misdiagnosis / diagnostic delay
• Understanding comorbid clusters
• Improving multi-specialty care

🎉 Final Summary in One Paragraph

Endometriosis is a common, chronic, genetically influenced disease where endometrium-like tissue grows outside the uterus. Modern genetics—especially GWAS—has identified dozens of risk regions, but linking them to true causal genes is hard because most variants regulate expression rather than protein structure, and these regulatory effects are subtle, tissue-specific, and cycle-dependent. By integrating GWAS with eQTL, splicing, methylation, and cell-type-specific data, researchers are beginning to map the regulatory pathways involved, revealing roles in reproductive development, epithelial regulation, proliferation, and inflammation. Although significant progress has been made, full biological understanding and translation to clinical solutions remain major challenges.