One of the biggest predictors of how successful a field will be is how well it fits the current tech stack. For example, AI blew up in 2022/23 because it came in the age where data was able to be stored in large amounts and processed. Without the data, or even the proper chips for that matter, AI would not nearly have had the same success. Another example of this is the internet, where we needed computers and signals (even this has evolved) for it to work properly. In short, until the prerequisite technology is developed to around 70% of it’s potential, it’s hard for technology to make progress.
In biotech there are several things that need to be solved before we enter an age of unprecedented growth. I’m not going to talk about these bottlenecks in this post but just for reference some of these include:
After fixing these things, the future of biotech will be an exciting place. We might even have to rename Silicon Valley to DNA Valley. While tech will still be needed, machine learning is already very advanced by this time so I think it’s safe to assume biotech will overpass it. In terms of a timeline, I think this shift will happen around 2040. So now that the scene is set, what fields of biotech are exploding right now?
Before we take a look at the future of biotech, it’s important to understand what we’ve already been able to do. This allows us to see which areas really need to most attention and what areas have been saturated. We can also learn valuable lessons from understanding what made these advancements successful.
Quite possibly one of the biggest advancements in biology is the discovery of DNA and genes. Watson, Crick, and Franklin were instrumental in this discovery that eventually led to us understanding the root cause of cancer, hereditary diseases, and syndromes. These discoveries will lead to the next generation of treatments where we can effectively target the root causes of disease. However, an often overlooked aspect of this discovery is GMOs. This term often has a negative connotation but it is definitely something that we depend on heavily in the farming sector. By modifying plants, we were able to get higher yields of crops while also making them healthier in regions that are nutrient deficient, eg. golden rice with higher vitamin A.
Even though there are still improvements to be made, we also have a very stable drug development process. Antibiotics were, and still are, one of the most important discoveries in human history. Drugs like statins help keep patients with cholesterol issues healthy and alive. This is all possible because of our understanding of how to engineer chemicals that interact with the body in (mostly) healthy ways. However, we also have strict regulations on the process with organizations like the FDA, which has saved many people from poorly designed treatments. Thalidomide is the primary example of this, where it was stopped from entering the U.S. due to its side affects and unstable evidence. This turned out to be a miracle as it was later shown that thalidomide was a dangerous drug that caused birth defects.
So now that we know what we know what has happened in the past, what does the future look like?
In the current phase of biotech, I would say this is the biggest sector. This field covers treating cancer, creating longevity drugs, curing diseases, and anything else that has implications in human health. However, just because it is where biotech has stayed for the longest time, it doesn’t mean that it’s unchanging. In fact, there are many shifts that are occurring right now that are changing the paradigms of what we used to think of as healthcare.
One of the more interesting shifts that will happen soon is the tailoring of treatments to a person’s genomic standpoint. Before this, healthcare stands on a “one treatment for all” basis. While we are all humans and share many characteristics, there are still some things that differentiate us. An example of this is our ability to metabolize drugs; we can be either rapid, normal, or slow metabolizers. This directly correlates to the success of drugs as the dose that’s meant for normal people can be too much for rapid metabolizers and too little for slow ones. By reading our genomic data, however, we are able to better understand how our body functions and design a treatment based on this.
The bottleneck of this future is the massive amount of data that would need to be analyzed for this to happen. For us to create individualized treatments for everyone, at an affordable price, we would need to have AI systems that can efficiently organize and deal with a patient’s genomic data. We haven’t reached model speeds that can do this yet, but once there is one, a whole field will be created almost overnight. A less thought of problem is the security risk of storing patient data. This is what made people hesitant of 23 and me’s product as they don’t know what people could do with access to their DNA. We need a surefire way of protecting patient data to have complete trust in these treatments.
Moving away from personalization, I’m also certain that we will start to see more complex therapies enter the market soon. CAR-T cell and CRISPR therapies have already been approved, showing the potential for these higher order, more advanced solutions. The use of existing systems in biology can offer a lot of potential for creating these treatments. However, this also has its own bottlenecks in that there is a really slow translation from the lab to the clinic. If we can fix this problem, we can see more innovative treatments get to patients at a faster rate.
As we shift towards more sustainable means of producing materials, there is no doubt that biology can play a big role in it. Organisms already produce many byproducts like carbon dioxide so what if we just manipulated them to produce products we need? This is a huge goal within bioengineering and comes down to a lot of factors.
One of the biggest problems with this process is understanding how a biological systems are interconnected and how these connections work. In current research, we like to only look at the results, eg. what does this protein do, what proteins are part of this pathway, what does this pathway do? Without understanding the fundamental principles behind how something works, it’s hard to build a complex solution to a problem. An analogy for this is planes. It’s possible to build a paper airplane without understanding how the forces of gravity and lift work, but it’s definitely not possible to build a full on passenger plane without understanding these principles.
To successfully engineer biological systems to produce materials that we want, we need to take a deeper look into either how biology works, or more systems found in nature. There’s always new species being found with new properties so if we get lucky, we may be able to find fungi that produce insulin, or bacteria that make concrete. However, I personally think understanding how biological systems work is a better solution to the problem as then, we would not need to rely on the luck of finding an organism that fits our needs. This could again be solved by machine learning or our own ideas too.
With the production of biomaterials, we need to start looking beyond products and instead try to engineer processes. This will give us a lot of control over our environment and potentially help a lot of people in the future. An example of engineering a process is making trees able to capture more carbon dioxide or produce more vitamin C. We have already seen part of this through GMOs in farming but the final frontier of this sector is a lot larger.
An example of a problem like this is growth time. Some products take a very long time to develop into the form we want, like pineapples which take on average 2 years to grow. To supply the world with its need for these products, we have to take up a lot of space so we always have the product in supply. However, a faster growth time engineered using biological principles could mean less waiting and less need for lots of space.
Another example of this type of problem is increasing production of a certain product. While producing a biosynthesized product is one thing, getting it to scale where it is economically viable is another thing. To achieve this, a lot of work needs to be put into maximizing the efficiency of the system so that it’s sole purpose can be to produce the substance. There are a lot of bottlenecks here but the main one is again, our lack of understanding of biology as a system.
I’ve mentioned machine learning a lot here so there is probably one question many people have: why not predict and simulate an entire cell? This is a problem that scientists have been working on for a long time and I would say is a very important task for the advancement of biology. If I were to pinpoint one thing in biology that could serve humanity the greatest in the realm of biotech, it would be the decentralization of research. However, biology is limited by its need for materials and experiments. Especially because of how expensive materials in the biotech field can be, this centralizes research to only big institutions and companies.
I think the way we fix this is by looking at how software has succeeded. Code is very decentralized; almost everyone can learn it and apply it with just a laptop. This is the primary reason why there are so many founders in the space that have made great products; they don’t need a PhD in computer science and 20 years of work experience to be in the position to build something great. For this reason, we’ve seen a rapid acceleration in the advancement of software. This is what we need in biology, a way to get research to everyone at a very cheap price (or even free).
The best way to do this would be to create fully functional digital cell. This would completely change everything as we could phase away from our need for physical materials to the point where everyone can perform experiments. My idea of this would be that the user could test something out, eg. a CRISPR knockout, point mutation, chemical treatment, etc. and then see its effects. This also brings the benefits of speed. We are no longer required to wait for cells to grow and for the changes to take place, as we could fast forward them with our digital version. Obviously, the more accurate the cell is, the more valid these experiments would be. If a person does find a therapy that works, then boom, they can just get easy funding for testing (because it has been accurately simulated using the digital cell) and start testing their therapy.
This is definitely a realistic future, but there are several obstacles that need to be overcome. The first of which is the insane amount of processing that would need to take place for this to happen. I think quantum computing could help us here, with improvements to the storage of data as well. The second bottleneck is making sure the accuracy of the cell is correct. We would need to do a lot of testing to make sure the simulations are running to best of their ability and at a high level of detail, making testing stage a long process. An idea here is to potentially build upon existing, proven ML models to verify the simulation in steps. However, once this is done, the opportunities here are endless.
I decided to separate this from disease treatment because of the timing at which I think these discoveries will come. While longevity has the characteristics of several diseases, it comes with its own unique challenges. The main challenge is variability. With a disease, its cause is often classifiable; a third copy of chromosome 21 leading to down syndrome, a point mutation in hemoglobin causing sickle cell anemia, and mutations in the HTT gene causing Huntington’s disease are examples of classifiable causes. With an understanding of the direct causes, it’s a little easier to track treatments based on the first principles thinking of directly targeting the root causes using CRIPSR, CAR-NK therapy, etc.
With aging, however, each person’s reason may be a little different. While characteristics like cellular senescence are spread out, the specific reasons why a cell becomes like this can be due to a multitude of factors. Even though we have senolytics that can increase lifespan and healthspan, this only treats the effects of aging, not its cause. While this is an ethical issue, I do believe that almost everyone wants to extend healthspan as we should have more quality time in life instead of losing precious time at the end of our lives to aging.
So what needs to be done so we can reach this step? Moving on from bioprocessing, where we need to understand the complex system of just one group of cells, we now need to understand how different systems overlap and interact with each other. This is a huge task that may border in the realm of needing energy and data capacity to scale efficiently. However, I think it is accomplishable and that we can get very close to understanding how our bodies work as an entire system at the most complex level.
Many people consider space exploration to be the final frontier of humanity’s accomplishments, a testament to how far we have come as a species. However, from a biological perspective, space exploration can be challenging due to two factors. The first one is usually the one most people can think of right away: time. Space travel takes a long time and there’s currently no ways for people to traverse these tremendous distances (such as to the next nearest star to us) in just one lifespan. One idea here is cryopreservation but we are not anywhere close to this while keeping a person still functional. We have a lot to work on and this requires us to understand the properties of biology at conditions unlike Earth. Proteins are molecular structures, and chemistry is often dictated by the external temperatures, making cryopreservation a difficult task.
This leads into the second reason, the effects of gravity on our body. Since the beginning of life, we have evolved to withstand gravity at a magnitude of 9.8N/kg. This includes all aspects of our cells: protein interactions, nutrient transportation, and the general flow of blood in our body. However, when we enter space, we suddenly don’t have those conditions anymore. Life has never evolved to withstand conditions outside of Earth’s gravity, so we don’t have a natural protection against it. Although concepts like creating artificial gravity through a spinning space shuttle are out there, this doesn’t help us when we have to traverse different planets with different gravity. We have to engineer our bodies to become resistant to these changes and withstand the changing gravitational force.
These concepts are what make up the synapse between physics and biology. We started with pure biology like understanding how blood flows and what cells are, then we went into biochemistry with DNA and proteins, until finally we go into biophysics, where we try to understand the applications of physics to solve complex biological problems. Since these problems are physics related, I believe that they will have physics based solutions.
An exciting field here is quantum biology. This is where we apply the concepts of quantum mechanics to biology to understand cells at a much deeper level. Hopefully, this can allow us to see the patterns in our systems and how we can engineer a solution to fix it. This is what I call the final frontier because of how complicated this would be. I would say this task is not in the range of machine learning at the moment. I think that we need a much more advanced method of ML to solve this problem, as current models have not performed very well with physics and math. Until then, humans can offer their insights by looking at existing information and trying to decode it with physics. If the math works out, I think we will be one step closer to solving this mystery and creating those solutions.
While all of these ideas present a great future for biology, it is also necessary to understand the risk associated with this type of research. With software, we obviously have many benefits like increased productivity, but we also have extreme downsides like social media addiction and new forms of harming people. The same goes for biology. For example, what would happen if someone used the digital cell to design a virus that perfectly crippled our bodies to the point of death. This would be a detriment to our survival as a species and it is important to always look out for the well-being of the community, as in the world.
Thank you for reading.
PS: feel free to reach out to me on LinkedIn if you want to discuss any of these topics