How generative biology can wipe out our most devastating diseases — for good

Biotech visionary Geoffrey von Maltzahn explains how deciphering patterns in our DNA helps untangle some of life’s great mysteries.

Compared to most people, Geoffrey von Maltzahn has a somewhat superhuman understanding of the untapped possibilities of science. Born into a long line of engineers, von Maltzahn was a kid when he began leveraging a combo of logic and DIY simple machines to make sense of the world around him. He went on to study chemical and biological engineering at MIT and Harvard and had already co-founded two companies focused on health-related innovation by the time he completed his doctorate. Today, he’s a partner at Flagship Pioneering, a firm dedicated to finding, funding and developing revolutionary biotechnology.

But despite his depth and breadth of knowledge, von Maltzahn says even he could not have anticipated how quickly machine learning would transform and expand the realm of life sciences. Six years ago, he co-founded Generate:Biomedicines, which he and his colleagues established with the intention of figuring out “the rules of proteins — the tiny sensors, motors and engines that make up life.” At the time, he says, it seemed like a stretch to view artificial intelligence as an especially useful tool in biology. But today, von Maltzahn continues, “we believe that all future protein therapeutics will be created with the help of generative AI.” The company is among the powerhouses on CNBC’s 2024 Disruptor 50 list, an accolade that reflects the growing importance of this technology in the modern medical landscape.

Here, von Maltzahn shines a light on the mechanics, challenges and limitless potential of generative biology.

So, what exactly is generative biology?

Generative biology is an emerging field at the intersection of AI and biology. With generative biology, we use machine intelligence and human intelligence to decipher the rules that govern the majesty of life and endeavor to generate new solutions for the health of people and the planet.

What does it mean to have a computer decode life — how does that generate biology?

DNA is a sequence of four letters: A, T, G and C. These four letters encode the quantitative function of every protein motor, machine, sensor — and ultimately the grandeur of life. While humankind has made amazing progress sequencing the code of DNA, that code has been more or less been gibberish to our brains. With the help of machine intelligence, we are starting to be able to truly read and write this code. This will enable us to generate new solutions for the health of people and the planet.

What don’t we currently know about the human genome that you’re itching to find out?

With every decade since the Human Genome Project, we are learning more and more about the human genome, with each wave opening doors to intervene in disease. For example, we assumed that we each had just one genome, but it turns out that we were off. All cells accumulate random genetic changes in their DNA, resulting in trillions of unique genomes in the body. Some genetic changes can make a cell resistant or vulnerable to disease; others can themselves cause disease. We founded Quotient Therapeutics in 2022 to study the vast genetic library and harness this new knowledge to develop the medicines of tomorrow. If our assumption about these new understandings of the genome is correct, it will open the door to many new ways to cure, prevent or reverse disease.

You’ve founded more than 10 companies and you lead a team of scientist entrepreneurs at Flagship. What drives you to do what you do?

Two answers to that question. First: A belief that the implications of biology transitioning from guesswork to predictable success will be profound — and that we are at the very beginning of this revolution. And second: Life is short and our scarcest resource is our time, so we should choose to do the things that seem irresistible and have the potential to change the world.

What are some key challenges in implementing generative biology for commercial use?

The goal of generative biology is to generate products that transform human health and sustainability. In the case of human health, that means new medicines, in which case the path to commercialization is the same — all potential new medicines, no matter how they’re discovered, go though the same clinical testing and regulatory review that we have today. In terms of scaling up platforms and companies that use generative biology, the key challenge is access to experimental data. We just saw this with OpenAI purchasing rights to the New York Times’s content archives: OpenAI needs data about language in order to improve its models and the results you get when you use ChatGPT. The same is true for biology — access to high-quality data is the key to scaling. We’re going to see lots of creativity around this soon.

If you could look into a crystal ball (or consult an algorithm) to predict the future, what do you think might unfold within generative biology in the next year or two? How about in five years?

Sometimes it is tempting to think that everything in nature has already been optimized because Mother Nature has had her lab open for a very long time. But if you take your average protein, you’ve got 100+ amino acids and 20 options each — that’s 20 to the power of 100-plus. So even with generous assumptions around everything that Mother Nature has ever done, she has only been able to test the equivalent of one drop of water versus all of the Earth’s oceans’ worth of combinatorial potential. What if tools like generative biology could help us explore vast new frontiers? I think the potential of this technology is extraordinary.

In the near future, we can imagine drug discovery moving from a lengthy, risky, trial-and-error endeavour to one that is predictable and fast. Generating new medicines in a fraction of the time and cost and with a much higher likelihood of success. Looking further out, I believe medicine will look a lot different than it does today. As medicine gains the ability to write the code of life in cells around our body, we’ll be able to confer vaccine-like protection against diseases like Alzheimer’s and heart disease. And we will be able to prevent disease from ever taking shape. A child with a single genetic mutation who might not see their second birthday today will, in the future, not remember the name of the disease they were born with.

Photo courtesy of Geoffrey von Maltzahn