TED Talks Daily · Society & Culture
What if you could read the DNA of your future child? | Noor Siddiqui
Noor Siddiqui, founder of reproductive startup Orchid, has developed the world's first technology to sequence the complete genome—all three billion base pairs—of IVF embryos, identifying risk for more than 3,000 hereditary diseases before birth. Siddiqui grew up watching her mother gradually lose her sight to retinitis pigmentosa, and that experience drove her to found the company. In her TED Talk she walks through how the technology works, shares real family cases, and raises unresolved questions about access inequality and the ethical boundaries of genetic selection.
Listen to this episode Go to the show page
What this episode answers
How is Orchid's whole-genome sequencing fundamentally different from conventional chromosomal testing?
Conventional testing only checks whether each of the 23 chromosomes is present—like confirming each chapter of a book exists—but never reads the letters inside; whole-genome sequencing reads all three billion base pairs and can detect variants as small as a few letters that determine whether a child will ever speak.
How widespread is the impact of hereditary disease worldwide?
400 million people worldwide live with hereditary diseases, half of them children, and while more than 3,000 such diseases have been identified, FDA-approved treatments exist for only 5% of them.
How much can parents actually reduce risk using polygenic risk scores?
With five IVF embryos, current models project a 30–50% reduction in risk; accuracy will improve as more data accumulates, promising even greater reductions from fewer embryos in the future.
Are there real cases of children born using this technology?
Yes—Lydia's parents, whose daughter has a severe neurodevelopmental disorder caused by a single-nucleotide variant in the KCNQ2 gene, used Orchid for their next pregnancy and their daughter Lucy was born without the same variant; separately, the mother of Mia, who has lissencephaly, screened her next embryo with Orchid and is currently pregnant.
How did Siddiqui respond to questions about cost and the ethics of how far genetic selection should go?
Siddiqui argues the cost is far lower than gene therapy (which runs $1 million per eye for retinitis pigmentosa), but on the question of equitable access and how far society should permit genetic selection, she said only that 'it is for society—including the TED audience—to decide,' leaving the answers to the audience.
The Era of the 'Genetic Lottery' Is Ending
At the moment of conception, egg meets sperm in a flash of zinc light and a new life begins. A mother carries two million eggs; a father more than 500 billion sperm. The odds that any specific pairing produces a particular person are one in a trillion—the same as picking one grain of sand from every beach on Earth. Yet hidden inside that miracle is the possibility of misfortune: more than 3,000 confirmed hereditary diseases that collectively affect 400 million people worldwide. Half of those patients are children, and 95% of those diseases have no FDA-approved treatment.
Orchid founder Noor Siddiqui grew up watching her mother lose her vision step by step to retinitis pigmentosa. It took years to reach a diagnosis, and the eventual verdict from doctors was that nothing could be done. "I cannot accept the reality that a random genetic typo means my mother will never see her grandchildren's faces"—that is the moment she traces her company's origin to.
The Critical Difference from Conventional Testing
Standard IVF genetic testing today checks chromosomal copy number: it confirms that each of the 23 "chapters" of the genome is present. Three copies of chromosome 21 signals Down syndrome. But it never reads the "letters" inside each chapter. A misspelling on chromosome 7 comes back simply as "chapter 7 looks fine."
Orchid's whole-genome sequencing reads every letter of all three billion base pairs. A change of just a few letters can be the difference between a child who will never speak and one who will—and the technology can detect it. The engineering challenge was the sheer scarcity of material: only five cells, containing 30 picograms (30 trillionths of a gram) of DNA, can be safely taken from an embryo. Orchid developed a proprietary amplification method that multiplies that sample hundreds to thousands of times without introducing contamination. Accuracy has been validated by matching embryo genomes against the genomes of babies born nine months later.
Cutting Risk by the Numbers
The screening covers two categories of disease. The first is single-gene disorders: cardiac defects, skeletal abnormalities, neurodevelopmental conditions, and hereditary forms of autism, among many others that conventional testing routinely misses.
The second is polygenic risk scores for complex diseases. Schizophrenia, bipolar disorder, and heart disease are each shaped by the cumulative effect of millions of variants. Models built from the genomes and medical records of hundreds of thousands of people can now quantify the risk for each individual embryo. With five embryos available today, a 30–50% reduction in risk is achievable. "The models we have now are the least accurate they will ever be," Siddiqui says. "They improve as data grows."
Real Families, Changed Outcomes
Mia was born with lissencephaly, a severe neurological condition that prevents most affected children from reaching their tenth birthday. Her mother had completed every standard carrier screening and NIPT available—tests that evaluate less than 1% of the genome—and none of them caught it. Mia's mother has since screened her next embryo with Orchid and is currently pregnant.
Lydia has a severe neurodevelopmental disorder caused by a single-nucleotide variant in the KCNQ2 gene. Her parents used Orchid when trying for a second child. Their daughter Lucy was born without that variant. "Same parents, same love, different outcome," Siddiqui says. "The difference made not by chance, but by choice."
The Potential—and the Open Questions
Genetic variants do not only bring disease risk; they also produce extreme phenotypes. A woman with Weaver syndrome standing over seven feet tall, a woman with achondroplasia under three feet—both differ from the human average by a single letter out of three billion. A Finnish Olympic skier carries a single-nucleotide variant in a red-blood-cell gene that gives him exceptional oxygen-carrying capacity. A family in the American Midwest has bone density eight times the normal level. A woman with a fourth color receptor can distinguish 100 million colors.
On cost, Siddiqui points out that gene therapy for retinitis pigmentosa runs $1 million per eye, making embryo screening far cheaper by comparison. But on whether the technology will be equally accessible to everyone, and on how far society should allow genetic selection to go, she offered no answer of her own. "That," she said, "is for society—including everyone in this room—to decide."
Takeaways
- Orchid's proprietary technology reads all three billion base pairs of an IVF embryo's genome from just five cells, enabling detection of risk for more than 3,000 hereditary diseases that conventional testing—which checks less than 1% of the genome—routinely misses.
- Polygenic risk scoring across five embryos can reduce the risk of conditions such as schizophrenia and heart disease by 30–50% today, with accuracy expected to improve continuously as more genomic and medical-record data is incorporated.
- The technology leaves two major questions unanswered: whether its benefits will be accessible equitably rather than concentrated among the wealthy, and where society should draw the ethical line on genetic selection.
This article is edited from a machine transcript of the first 30 minutes of the episode. It is not an official summary — please listen to the episode itself.