OpenAI与波士顿儿童医院合作：AI助力罕见病诊断

Welcome everyone. Thank you for joining us again. Uh today's conversation looks at a research collaboration between the Manton Center for Orphan Disease uh orphan Disease Research at Boston Children's Hospital and Harvard and ourselves at OpenAI. Um, so you should know that hundreds of millions of families worldwide struggle with rare disease and the average time just to get a diagnosis is six to seven years. So it's it's a huge long struggle for them.

Um, and this research is super important. So in this in this study across 376 cases, an AIdriven workflow surfaced evidence linkeds that led to 18 diagnoses of rare disease. So we want to dig into that. First, we're going to hear from Stavronis, a Maten Center research participant, about his search for a diagnosis for his own rare condition and the impact of getting diagnosed. We're then going to hear from three study co-authors.

Dr. Katherryn Brownstein is the scientific director of the Maten C Center's gene discovery corps where she focuses on genome analysis and barriers to diagnosis. Dr. Alan Begs is director of the Maten Center and brings decades of experience in human genetics and rare disease. and Suya Shringerpur is a machine learning researcher and a statistical geneticist at OpenAI working across AI, genomics and computational biology.

So they'll discuss why rare disease is still hard, how this workflow operated, what the study found, and what its limitations were. Then we'll open up the conversation to the community. So let's start with Stav. St. Hi, welcome.

Thank you. So stop, you first noticed symptoms as a teenager, but you spent years searching for the right diagnosis, including an initial diagnosis that turned out to be wrong. Can you tell us first about yourself? Um what where do you sit? What do you do? And then we'll get into the story of of your search for diagnosis.

Yeah. Yeah, absolutely. Uh thanks for having me. So my name is Stav Rones. Uh was born and raised in the Boston area. Um, growing up, uh, you know, I always, uh, loved playing sports, different sports, baseball, basketball, rowing, and, um, when I was, uh, kind of getting more competitive in my teenage years, I started to notice that I was having certain issues that other kids were not facing. Um, and it was pretty frustrating not knowing what the issue was.

I was getting certain pains, limitations with movements. Um, and you know, eventually tried to to seek professional medical help. And um it took a long time to even figure out that it may be genetic related. Um and then to get from even there to doing a full genome testing and figuring out what kind of uh rare disease it was was a was another whole process. But uh currently uh yeah I'm based out of Miami, Florida and uh I'm a software engineer but um yeah.

Yeah. So so started when you were a teenager. How long did it take you to get that the correct diagnosis? It took a long time. Um I mean there are a variety of different factors you know like trying to correlate certain physical issues with having a genetic a rare especially a rare genetic disease right. Um I mean certain things express themselves that are common diseases it's more easy to diagnose but with rare diseases because there's so little patients and so little understanding about what the how it presents itself um what kind of symptoms it took at least 3 to four years to figure out that it wasn't just a physical issue that that it was neurological um and and from knowing that it was neurological to get to the point that it was genetic was a wasful process at least five years.

Yeah. Wow. Um, so you got your genome sequenced and then you got a diagnosis and you got clarity. What did that unlock for you? What was life like? Uh, and what were you able to do after you got clarity?

Yeah, absolutely. I mean just in terms of um you know the the knowing exactly what the issue is um just kind of helps you situate yourself and and um adapt to to knowing what you have because if you don't know what the issue is, you know, you might do things that are harmful. um and knowing exactly what you have to um you can be part of a community of others with the same thing and that's um you know it's just a really nice to to kind of understand and and be with others who have the same thing.

Um and also really importantly um there are um companies and there are teams working on cures and solutions for these rare diseases. So knowing exactly what you have, you can actually um start to get involved and potentially uh find you know find cures uh for the river disease. And then also um for me the biggest things is when it came to family planning. So knowing the implications of um what having the disease does in terms of uh when you go to have children uh was very important for me and um using modern technology how you can kind of um you know get around that.

Yeah, that's amazing. Thank you for sharing that story. Um, we're going to bring on the the three study co-authors we have here, Kathern Brownstein, Alan Begs, and Su Shringerpur. Um, to kind of widen what we're uh what we're discussing here. So, I'd like to start the panel with Katherine Allen Su. Um, Allan, I I'd like to start with you first. St. Search for an answer took years. How common is an experience like his and why can rare disease cases remain unsolved for so long?

Sure. Well, it unfortunately it's way too common. People often refer to this as a diagnostic odyssey. Um folks know that something's going on. In many cases now in recent decades, we've realized that many of these rare conditions have a genetic basis. But our ability to uh test all the genes that we have been very difficult. To put the problem in context, our genome, the DNA that we inherit from each of our parents is roughly three billion bases.

And those three billion bases encode something in the order of 20,000 different genes. And then each gene can have many different ways that it is damaged. And we know now of the number is constantly going up, but but between eight and 9,000 different genes that are known to be associated with disease when they're abnormal. Um and so patients in the past we only had the ability to look at a few genes at a time. What's changed is newer technology generically called next generation sequencing that allows us to sequence everything.

But then our data sets of course contain upwards of hundreds of thousands to millions if not more genetic data points. And uh as we'll get into what the ai allows us to do is help hone in on the relevant ones. So this has been for many families a long process this so-called genetic or diagnostic odyssey. Um and through uh collaboration with research institutes like us and other around the world uh people eventually get an answer.

Yeah. So you're really looking for needles in haststacks.

Yeah. That's what people often say. Yeah. Um, thank you, Katherine. Um, you said the bottleneck is time because geneticists are sorting through varants and databases and complex ambiguous symptoms and new research. What does that time bottleneck look like for you when you're working on an unsolved case? So when I don't know when no one knows what's going on, you try to approach a case with an open mind and really kind of start at the beginning and look at the known genes, see if something was missed.

Um then you broaden it out and you if you have a complete trio that is mother and father and child, you look at what's not in the parents but they're it but is in the child. So, it's an error during development or at some point along. And all of us have these things that are called denovo mutations that aren't in either of our parents, but they're in us. And there it's a good place to start to kind of look and see, okay, is that what's going on here?

Is there was a denovo variant that could be disease-causing? or you kind you can look at things where there's each parent is a carrier and the child got two uh variants, one from each parent. U

原文超出正文长度上限，此处截断——上游还有内容，完整版见上方「原文 ↗」。