Imagine a world where a simple drop of blood, a finger-prick, or a wearable sensor provides a continuous readout of your biological health – without having to visit the hospital. Where AI-powered tools analyze not just your DNA, but your entire molecular profile, to predict risks, guide personalized treatments, and even prevent diseases from developing at all. What would it take to make this vision a reality? What role could analytical science play? And can we get there by 2050?
Here, as part of a series of articles tackling these important questions, we speak with Mike Snyder, Mike Snyder, Stanford W. Ascherman Professor of Genetics, Director of the Center for Genomics and Personalized Medicine, Stanford School of Medicine, USA, who believes that a combination of biochemical data, imaging, and real-time tracking will be incredibly powerful for early disease detection, prevention, and personalized treatment.
Has the Human Genome Project lived up to the early expectations?
I think the Human Genome Project held great promise, and in my opinion, it has delivered – though perhaps more slowly than many expected. However, the fundamental idea that we can sequence human genomes and make risk predictions has proven to be valid.
Of course, we now understand that genetics accounts for only about 16 percent of lifespan – there are many other factors beyond genetics that influence health outcomes. But the fact that we can sequence someone’s genome and make risk predictions is still incredibly valuable.
In my own case, for example, my genome predicted that I was at high risk for type 2 diabetes. As some readers may know, I conduct extensive deep-data profiling on myself and a group of individuals. During this process, I actually developed diabetes. Because my genome had indicated a higher risk, I was already on alert, which allowed me to catch it early and get it under control.
Beyond individual cases, the Human Genome Project has been foundational for basic research. Our understanding of biology and human health has expanded dramatically in ways that wouldn’t have been possible without having the genome sequence as a reference. But it is just the tip of the iceberg. The genome provides useful information, but it doesn’t tell the whole story. That’s why we need additional layers of measurement – beyond genetics – to create a more comprehensive profile of human health. Only by integrating these other data types can we truly enable personalized medicine and treatments.
<4>What are the main health challenges that could – or should – be tackled with precision or personalized medicine?
I would say it applies to everything. We should be able to build predictive models to assess individual risks for various conditions. I like to think of health and disease as being influenced by many factors – your DNA is one of them, and an important one, but it’s only part of the picture. Environmental factors, lifestyle, diet, activity levels – all of these also play a crucial role.
We should be able to build personalized models that incorporate all these elements. Our lab is doing a lot of work in this space, particularly in glucose control. It turns out that glucose responses to food are highly personal – some people spike to bread, others to pasta, others to potatoes. The microbiome plays a role, as does genetics, but neither tells the full story.
This is a critical issue because in the US, 11.6 percent of the population has diabetes, and 33 percent are prediabetic, most of whom will eventually develop diabetes. Understanding glucose regulation on an individual level is crucial because it varies so much from person to person. If you can identify which foods spike your glucose and what factors help mitigate those spikes – whether it's specific foods or physical activity – you can create personalized strategies to maintain healthy glucose levels.
We’ve already applied this concept in practice. I co-founded January AI, a company that builds personalized predictive models for glucose regulation. Other companies are working in this space as well, and the results are promising. People who adopt these personalized approaches often lose weight and improve their metabolic health. If you’ve ever worn a continuous glucose monitor (CGM), you’ll quickly realize how it changes your eating habits – it’s hard to ignore the data when you can see in real-time how different foods affect you.
Beyond glucose monitoring, wearables like smartwatches are another powerful tool. They can track heart rate continuously and detect when someone is getting sick – often before symptoms appear. We’re now in a world where we can detect early signs of illness remotely.
Proteomics is another important window into health. While genomics and transcriptomics provide valuable insights, proteins and metabolites are often much closer to an individual's actual health state. Advances in proteomics have been driven largely by mass spectrometry and capture-agent-based technologies, and now the combination of both is extremely powerful for profiling health.
What’s even more exciting is that remote health monitoring is becoming increasingly feasible. We’ve developed microsampling assays that allow people to provide just a tiny drop of blood – about 10 microliters – from a finger prick, which can then be mailed in for analysis. From that one drop, we can now profile 7,000 analytes, including proteins, metabolites, and lipids.
This is the foundation of Iollo, another company we spun off. We can analyze these blood samples and provide insights on 600 metabolites, which reflect various health areas like oxidative stress, inflammation, and heart health. It’s a simple, scalable way to provide in-depth health profiling remotely.
On the disease side, precision medicine is already making a major impact. Lung cancer is a great example – it’s incredibly heterogeneous, and treatment depends heavily on genotyping the tumor. If you get lung cancer, getting your tumor sequenced is crucial because your treatment plan will be determined by the genetic mutations in your cancer cells.
So yes, I think precision medicine has applications across the board, from metabolic health to early disease detection to cancer treatment. The technology is advancing rapidly, and we’re just beginning to see its full potential.
Do you see a future where people regularly measure their proteome – or even broader biometrics – using micro-sampling to get a baseline reading?
Absolutely, and it wouldn’t just be the proteome – it would include metabolites and other markers as well. I think this will be incredibly important.
Right now, most people don’t measure their health regularly. They wait until they’re sick, then they get tested, and the system shifts into treatment mode. But what’s truly powerful is tracking health while people are still well. Establishing a baseline profile is critical because it allows us to detect subtle shifts over time – before a serious issue develops.
A great example of this is whole-body MRI. I strongly believe in comprehensive health measurements, and if you ask most physicians whether everyone should get whole-body MRIs, they’ll say no – because you’re going to find nodules. Women commonly have them in their breasts, men in their pancreas or prostate, and so on. Doctors worry about false positives leading to unnecessary procedures. But if you ask me, I’d say absolutely, everyone should get a whole-body MRI. The issue isn’t whether you have nodules – most people do. The real question is: Are they growing? That’s something you can only determine by tracking a person’s health over time.
This same principle applies to proteomics and metabolomics. By establishing a baseline for each person, we can detect when something deviates from the norm, allowing for early intervention. That’s why we believe so strongly in deep data profiling – tracking a person’s normal state so that when something does shift, we can identify it early and respond accordingly.
Now, coming back to micro-sampling, its real power lies in convenience. Most people don’t want to go to the doctor unless absolutely necessary – it’s time-consuming and inconvenient. But micro-sampling allows people to collect small blood samples at home – from the shoulder, fingertip, or another site – then mail them in for analysis.
These samples can be used to measure far more biomarkers than what’s typically tested during a routine doctor’s visit. Plus, they provide a more accurate reflection of real-life health, because people are more relaxed at home compared to a clinical setting, where anxiety can cause spikes in heart rate, blood pressure, and other metrics.
So yes, I think micro-sampling, combined with longitudinal health tracking, is going to be a major shift in how we monitor and manage health in the future.
Do you think the technology is already advanced enough for large-scale implementation?
I think we could do this right now. The technology is there, but you’d need to go through all the necessary regulatory steps to demonstrate that the protocols are robust, reproducible, and validated as diagnostic assays. That’s a process, but it’s feasible.
The bigger issue, though, is who is going to pay for this? At least in the US healthcare system, the problem is that it’s highly fragmented. On average, people change their healthcare provider every 18 months, which means no single provider has a long-term financial incentive to invest in preventative care. Why spend money on proactive health monitoring if the patient will be someone else’s responsibility in a year and a half? That’s a major challenge. Ideally, we should be moving toward a system where people get profiled while they’re still healthy, and their health plan covers it. But right now, the financial incentives aren’t aligned.
That’s what needs to change. We need to build financial incentives that encourage preventative health monitoring. For example, I believe people should get paid to wear a smartwatch that tracks their vitals. They should receive discounts on their health plans if they participate in routine micro-sampling every three to six months. This would allow for early detection of health issues, making treatment more effective and far less expensive than waiting until someone is seriously ill.
Right now, we’re reactive – we wait for problems to develop and then treat them. But with the right incentives, we could shift toward proactive health management, ultimately reducing costs and improving health outcomes.
Is there anything still missing from the analytical toolbox to make this vision fully viable?
Although the core technology exists, there’s still a lot of validation work needed – especially around reproducibility and clinical integration.
One of the biggest challenges is that we’re collecting more data than we fully understand how to use in a clinical setting. The information we gather is valuable, but we still need to refine how we interpret and integrate it into medical decision-making. For example, with micro-sampling, we’ve shown that we can measure incretins, insulin, and other important biomarkers, which provide useful insights. However, ensuring that all these assays are quantitative, reliable, and standardized is essential.
Beyond that, the medical system is still built around a limited number of biomarkers. When you go to a doctor today, they typically track only about 15 key markers – a relatively small number. But now, with high-throughput technologies, we can measure hundreds or even thousands of molecules in a single sample. The challenge is determining how to use this high-resolution data effectively – both to establish a healthy baseline and to better classify diseases.
A great example is diabetes. Right now, we categorize it broadly into type 1, type 2, and MODY (maturity-onset diabetes of the young). But in reality, type 2 diabetes is likely made up of 50 different subtypes. With deeper molecular profiling, I’m confident we’ll be able to subtype diseases more accurately, leading to more targeted and effective treatments for each individual.
And ideally, a lot of this can be managed through lifestyle adjustments. Take glucose regulation, for example – if you know which foods spike your blood sugar, you can adjust your diet accordingly to prevent issues before they develop. Personalized insights like this will allow people to take control of their health, ultimately reducing disease risk and improving long-term outcomes.
Do you see a role for sensor technology that bypasses the need to send samples to a lab?
Absolutely – that’s coming. We already have a continuous glucose monitor (CGM), and soon, we’ll likely see wearable sensors for other markers like ketones and lactate. I think many people would love to have continuous monitoring for cytokines, key proteins, and biomarkers for major health events – for example, troponin for heart attacks or biomarkers for Alzheimer’s disease.
Setting up these assays takes time, but it’s definitely doable. The real challenge will be bringing the cost down so that people actually adopt these technologies. Take Alzheimer’s disease, for instance – you probably wouldn’t need continuous monitoring since it progresses slowly. But for faster-developing conditions – such as certain cancers – having continuous monitoring could be extremely valuable.
Cytokine tracking would also be very powerful because inflammation is linked to almost every major disease. A continuous cytokine monitor could help detect early signs of disease or flare-ups in conditions like autoimmune disorders, cardiovascular issues, or infections. If something spiked, you’d know to investigate further and take preventive action.
So yes, wearable sensor technology is going to be a huge area of development, and it will fundamentally change how we track and manage health in real time.
Putting the pieces together, what is your vision for the future of precision medicine?
When it comes to biochemical testing, I see a combination of different sampling frequencies emerging. There will be continuous monitoring – this is useful for tracking specific biomarkers in real-time, like glucose, cytokines, or lactate. Then there’s frequent but less invasive sampling – finger-prick devices that instantly analyze a small panel of key markers (maybe half a dozen or up to 20 analytes), done weekly to track overall health trends. And finally, deep profiling at longer intervals will be key – micro-sampling done every few months to get a very comprehensive biochemical profile, measuring thousands of molecules at once. This would provide high-resolution insights into multiple wellness categories, disease risks, and metabolic health. The deep sampling would be particularly powerful because it gives a broad view of your health, rather than just focusing on one or two markers at a time like continuous monitors tend to do.
So, I think the future of precision medicine will involve a combination of these approaches – continuous tracking for short-term trends, frequent finger-prick tests for mid-range monitoring, and in-depth molecular profiling at regular intervals to provide a comprehensive view of health over time.
You’ve mentioned that cost and payment structures are major barriers to implementing precision medicine at scale. Are there any other significant hurdles?
A lot of people worry about privacy, though personally, I don’t see it as a major issue. The bigger challenges are navigating the regulatory process and validating these tests to demonstrate their clinical utility.
I think the way these technologies will initially roll out is similar to how genome sequencing started – it will begin as a concierge service, where wealthier individuals can pay out of pocket for early access. This is already happening with services like Iollo and January AI, where people can order personalized health insights, but they have to pay for it themselves.
The key next step is proving the value of these tools – showing that they provide actionable health insights that lead to better outcomes. Once that happens, the hope is that they will be covered by insurance or adopted by national healthcare providers, making them accessible to everyone, rather than just those who can afford to pay privately.
Ultimately, cost and healthcare system adoption will determine how quickly these innovations become mainstream.
Any additional thoughts on the role of analytical and measurement science in making precision medicine a reality, particularly in the near term?
One key area is ensuring that assays are truly quantitative. Right now, particularly in discovery research, we often rely on relative measurements, such as mass spec ion counts, rather than absolute concentrations. But for clinical applications, absolute quantification is essential. If I get tested today and then again in three months, I need precise and comparable measurements to track changes over time. Many assays aren’t set up that way yet, but they need to be.
Another critical factor is throughput. For some assays, scalability is still a bottleneck. Mass spectrometry, for example, is an incredibly powerful tool, but even a targeted assay can take 20 minutes per sample. Meanwhile, other high-throughput platforms can analyze thousands of samples in the same time frame if they’re properly parallelized.
So, increasing throughput and automation will be crucial, especially if we want to scale precision medicine to handle millions – or even billions – of tests per year, like those processed by major diagnostic labs.
Would you say you’re generally positive about the future of precision medicine?
Absolutely – this is the future.
Some areas will roll out faster than others. Wearables, for example, will likely expand quickly because they’re affordable, easy to use, and provide real-time insights. Micro-sampling will also gain traction relatively soon, as long as it continues to demonstrate clinical utility.
So yes, I’m very optimistic. I think the world will look very different in 10 to 15 years – or at least, I hope so. That’s exactly what our lab is working toward – measuring everything we can to help keep people healthy for as long as possible.
Newsletters
Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.
