How AI Could Help Us Understand Hypermobile Ehlers-Danlos Syndrome

One of the topics I’m most passionate about - and one that I’d like to share with you today - is Hypermobile Ehlers-Danlos Syndrome (h-EDS) and how AI tools might be able to revolutionise the research being done to understand it.

What is h-EDS?

Hypermobile Ehlers-Danlos Syndrome is a chronic, often invisible condition that weakens the body’s connective tissue -the biological "glue" that holds everything together. This means it can cause widespread pain, fatigue, and joint instability. But it can also affect organ tissue, skin, digestion, and even brain function.

Despite its wideranging impact, there is no known genetic marker for h-EDS, and many people go years - sometimes decades - without a proper diagnosis. Because it’s invisible both to the eye and often to current scientific tools, h-EDS remains misunderstood and frequently misdiagnosed, even though it may affect as many as 1 in 500 people.

There are 13 recognised types of Ehlers-Danlos Syndrome, and h-EDS is the only one without a known genetic mutation. This mystery is exactly what draws me to it.

Why is it so hard to research?

Even though researchers are working hard to identify the genetic basis of h-EDS, it’s incredibly difficult because there’s no single, obvious genetic mutation behind it—unlike other types of EDS, which are linked to specific genes.

Scientists now think h-EDS is likely polygenic, meaning it’s caused by a combination of multiple gene variants, each contributing in a small way. To make it even more complicated, symptoms vary widely between individuals. h-EDS isn’t a straightforward condition. Symptoms can be represented on a spectrum, but it isn't a linear spectrum it’s more like a circular spectrum as seen in the image below

Different people have different “slices” of symptoms: joint instability, fatigue, brain fog, chronic pain, digestive issues, and more.  That means that no two people experience h-EDS in the exact same way - and that makes it really hard to research using traditional tools.

How genetic research is currently done

Right now, researchers often rely on whole genome sequencing (which reads every base pair of DNA) or exome sequencing (which looks only at the protein-coding 2% of DNA). They compare the DNA of patients with h-EDS to healthy controls to try to find patterns.

But this is a slow and limited process, and it relies heavily on manual comparison and statistical analysis. With complex conditions like h-EDS where the differences between patients are hugeit ’s extremely difficult to draw meaningful conclusions with small sample sizes and traditional tools.

This is where AI comes in.

You know what could help us find those subtle, complex patterns? Artificial intelligence - and in fact, it already is!

Machine learning, a form of AI, can analyse thousands of genomes at once, spotting tiny combinations of genetic variants that may correlate with h-EDS symptoms. These models can detect patterns that are nearly impossible for humans to see - especially when each variant has only a tiny effect on its own.

But AI’s potential goes beyond just gene analysis.

It can combine genetic data with so much clinical information like symptom history, MRI scans, joint flexibility scores, and more to find deeper, previously unseen connections. It could even help classify new subtypes of h-EDS based on patterns of symptoms and genetic markers. This would be a massive breakthrough - it could lead to more personalised treatments and help patients understand their condition better.

Even more powerfully, AI tools can be trained on known cases to flag patients likely to have h-EDS long before a human doctor might recognise it. Given how often h-EDS is misdiagnosed (trust me it’s happened to me more than once), this could be life-changing.

Why this matters to me

I became interested in rare diseases and chronic illnesses like h-EDS because I have h-EDS. This makes it so much more real to me and it's why I realise how important it is an how much it can affect day to day life. But the reason it matters to me is more than just the affect it has on me. To me, science is about solving puzzles that matter, and I believe one of the most meaningful "puzzles" we face is helping people with invisible illnesses feel seen, understood, and supported.

We may not have all the answers yet, but I’m hopeful that by combining genetics, data, and powerful tools like AI, we’ll begin to unlock the mysteries of conditions like h-EDS and help thousands of people who deserve clarity, care, and compassion.

Sources:

Elhers Danlos Society

Washington Post Live - https://www.washingtonpost.com/washington-post-live/2025/02/27/transcript-rare-diseases/

Mayo Clinic - https://connect.mayoclinic.org/blog/ehlers-danlos-syndrome/newsfeed-post/can-artificial-intelligence-be-useful-for-eds-and-hsd/

NIH - https://www.ncbi.nlm.nih.gov/books/NBK584966/