When Wei-Wu He, PhD, executive chairman of Human Longevity Inc. (HLI), asked the company’s AI platform how long he could expect to live, it gave him an estimate of 85 to 90 years. Unsatisfied, he asked what he would need to change to reach 95. The system identified five parameters, including his metabolic health, that could potentially extend his predicted lifespan. He then asked how much additional time he might gain by improving his blood glucose regulation. The answer: two years.
Whether such predictions will ultimately prove accurate remains an open question. But for He, the exchange illustrates the next frontier of precision medicine: moving beyond identifying an individual’s genetic predisposition to disease toward predicting how clinical interventions might alter their long-term health trajectory.
The ambition builds on a mission established by genomic pioneer Craig Venter, PhD, who co-founded HLI in 2013 to investigate the relationships between inherited genetic variation, observable physiological characteristics, and human longevity. More than a decade later, the company has assembled a longitudinal dataset integrating whole-genome sequencing, whole-body imaging, blood-based biomarkers, and clinical information from thousands of individuals. HLI is developing AI models that combine this genetic and phenotypic data to identify individuals at elevated risk for cardiovascular disease, cancer, dementia, and metabolic disorders, with the longer-term goal of informing personalized preventive interventions.
In this interview with Inside Precision Medicine, He discusses the scientific and clinical challenges of translating genomic risk prediction into improved health outcomes, including the limited generalizability of existing algorithms across genetically diverse populations. He also explores the potential of blood-based diagnostics to expand access to precision medicine, the emerging therapeutic possibilities of genome editing, and the ethical implications of incorporating genetic information into reproductive healthcare.
This interview has been edited for length and clarity.
He: They overlap, but they’re not exactly the same thing. Think about the human body as the most extraordinary machine on Earth. It’s incredibly complex, yet it can function for 80, 90, or even 100 years. The basic code for that machine is your genome. You inherit approximately three billion DNA base pairs from each parent. That information directs the development of your body from a single cell. A mutation might contribute to the development of a tumor, or other changes might interfere with normal cellular function.
I think the world still underestimates how powerful the information encoded in our genome really is. The revolution is that sequencing the first human genome cost billions of dollars. Today, genome sequencing is available for a few hundred dollars. And you generally only need to sequence your inherited genome once in your lifetime.
IPM: Given what you just said about the information theory of aging, wouldn’t you want to sequence someone’s genome more than once to capture the changes that occur over time?
He: That’s an important distinction. Your inherited genome is largely stable, but your somatic genome can change as you age. For example, a tumor can acquire mutations and shed DNA into the bloodstream. If you sequence that circulating DNA deeply enough, you may be able to detect those mutations. That’s the principle behind liquid biopsy. There are different approaches, including methylation-based testing and tests that detect tumor-associated DNA mutations.
So, while you generally only need to sequence your inherited genome once, there are other types of molecular testing that can provide important information as you age.
IPM: What exactly is Human Longevity offering today, and how has the company’s approach evolved since its founding?
He: Our original mission was to link genotype with phenotype. Your genotype is the genetic information you inherit from your parents. Your phenotype includes the observable characteristics and health conditions that develop throughout your life, such as cardiovascular disease, cancer, or changes in your eyesight. We want to understand how the two are connected.
When we started, collecting that information was extremely expensive. Whole-genome sequencing alone cost us almost $10,000 per person. We also collected whole-body MRI scans, metabolomic and proteomic data, and traditional blood panels. Our initial membership was expensive, so it was accessible primarily to wealthy individuals.
I compare that to Tesla’s original Roadster. It was an expensive product produced in limited numbers, but it helped establish the technology that eventually made more affordable vehicles possible. As sequencing costs have fallen, we’ve been able to introduce a much more affordable product.
What distinguishes our approach is the data behind our application. We’ve spent years collecting extensive clinical and molecular information from our clients, including longitudinal phenotypic data. Sequencing itself is becoming a commodity. The real value is increasingly in interpreting the genome and connecting it to meaningful health outcomes.
IPM: Beyond whole-genome sequencing, you offer whole-body MRI and other diagnostic tests. Is one modality more valuable than the others?
He: I don’t think there’s a single test that’s better than all the others. There are thousands of human diseases, and everybody’s risk is different. Some rare genetic diseases affect only a few hundred people worldwide, but if you’re one of those people, identifying the relevant mutation is enormously important. No single test can capture everything.
I would argue that whole-genome sequencing should be the foundation because it’s becoming so inexpensive. But you need to combine it with other information, and you need AI to help interpret the results. In the future, genome sequencing itself will be a commodity. The analysis of the genome will not be.
IPM: Whole-body MRI is considerably more expensive than genome sequencing. What distinguishes Human Longevity’s imaging approach from other companies offering similar services?
He: There are cheaper whole-body MRI services available, but there can be significant differences in how the scans are performed. We use a Siemens MAGNETOM Vida 3-tesla MRI system, and our examination takes approximately an hour and 20 minutes.
We spend additional time imaging specific organs, including the pancreas, because we want to identify abnormalities as early as possible. Different organs require different imaging protocols. A shorter examination may not provide the same level of detail across every organ. Consumers often see two companies offering whole-body MRI and assume they’re purchasing the same service. But the equipment, imaging protocols, and interpretation can be very different.
The same applies to whole-genome sequencing. Two companies might sequence the same genome, but the quality and depth of their analysis can differ considerably. That’s why we’re investing so heavily in the algorithms behind our clinical platform.
IPM: You’ve described how AI can combine genomic, imaging, and longitudinal clinical data. Can your application also predict how long someone might live and identify changes that could improve that prediction?
He: That’s exactly what we’re working toward. I recently asked our application how long I was likely to live. It predicted that I would live to approximately 85 to 90 years old. I then asked it to identify five parameters I could change if I wanted to live to 95. One of the areas it identified was my metabolic health. I asked what would happen if I improved my hemoglobin A1c and insulin resistance.
That’s the power of a neural network. When you change one parameter, the predicted outcome can change. And the model becomes increasingly informative as we incorporate additional clinical data.
IPM: How do we democratize this approach? If these comprehensive assessments are primarily available to wealthy people in the United States, aren’t we at risk of widening existing health disparities, particularly in populations that are already underrepresented in genomic research?
He: Every major technology goes through a process of diffusion. Think about the light bulb. Initially, it was expensive and available to relatively few people. Today, it’s an inexpensive technology used around the world. Genome sequencing is following a similar trajectory. We’ve gone from spending billions of dollars to sequence the first human genome to offering sequencing for a few hundred dollars.
AI is making it possible to interpret that information on a scale that would have been extremely difficult before. I believe we’re entering a new revolution built around information and neural networks. In the future, we might collect 20 terabytes of information about an individual, and a neural network could use that information to answer increasingly sophisticated questions about their health. But you don’t need to begin with 20 terabytes of data. You can start with the genome.
At a sufficiently large scale, sequencing costs could fall further. A country could potentially license our software and offer genomic risk assessment to millions of people at a relatively low cost. That’s how I envision democratizing this technology.
IPM: Would whole-body MRI eventually become part of that more affordable model, or are there other diagnostic modalities you would prioritize?
He: Whole-body MRI would probably be one of the last things I’d include in a democratized version because of its cost. The second step after genome sequencing would be a blood test. If you collect a tube of blood and measure approximately 300 biomarkers, you already have an enormous amount of useful information. From there, you can incorporate more comprehensive proteomic measurements and other clinical data. The goal is to build the platform incrementally, starting with the most accessible and informative measurements.
IPM: You were already leading Human Longevity before Craig Venter passed away. How much of the company’s current direction reflects his original vision, and where are you putting your own stamp on it?
He: The mission has always been the same. When I asked Craig why he called the company Human Longevity, he explained that “human” meant all eight billion people on Earth. Longevity was about using science to extend healthy human life. We believe that humans have the biological potential to live beyond 100 years, and we want to use science to help more people reach that potential while remaining functional and healthy.
But there’s another equally important part of the mission: shortening what I call the “sick span.” Americans spend many years living with illness, and that places an enormous burden on individuals, families, and the healthcare system. If we could dramatically shorten the period of illness near the end of life, we could improve people’s quality of life while potentially reducing healthcare costs. That’s particularly important as the global population ages.
The question isn’t simply how to extend human life. It’s how to prevent people from spending those additional years living with serious disease.
IPM: What would that look like in practice?
He: Osteoporosis is a good example. Imagine sequencing someone’s genome when they’re young and identifying an elevated genetic risk of osteoporosis or bone fracture. You could then monitor their bone density and relevant clinical biomarkers throughout their life.
If their bone density begins to decline, you could intervene before they experience a fracture. Their health application could alert them to their elevated risk and recommend appropriate preventive measures. That’s the future of medicine: identifying individual risks early and making relatively small interventions over time to prevent serious problems later.
We’re also speaking with insurance companies about how this approach might be incorporated into their products. If insurance can help us democratize access to this technology, that’s something we’re interested in exploring.
IPM: Could we eventually reach a point where someone has their genome sequenced at birth, receives an estimated lifespan, and sees how that prediction changes throughout their life?
He: In some ways, that’s already happening. There are inherited genetic conditions that can substantially affect life expectancy. If you sequence someone’s genome and identify a serious genetic disorder, you may already have information about their likely health trajectory. But the exciting thing is that we’re beginning to develop therapies that can change those trajectories. Take sickle cell disease. We now have an FDA-approved CRISPR-based therapy. We don’t yet have the long-term data to know exactly how much that treatment will extend patients’ lives, but I think the potential is extraordinary.
That’s the revolution: we’re not simply identifying genetic risks. We’re beginning to develop technologies that may allow us to change their consequences.
IPM: As genomic medicine becomes more widely available, it also raises questions about how genetic information might influence reproductive decisions and public policy.
He: We’re already seeing examples of that. In countries such as the United Arab Emirates, genetic screening is being incorporated into premarital healthcare programs to identify couples who may carry mutations associated with the same inherited disorders. The goal is to identify reproductive risks and provide couples with information about their options, including assisted reproductive technologies. I think this demonstrates how genomic information is beginning to influence healthcare at a societal level.
Genetic screening is already becoming part of healthcare systems, and I think we’re going to see much more of it.
