Can you imagine a world where our organs are grown for transplant rather than harvested from car crash victims, where vaccines can be developed and delivered in weeks, and where we can actively work to create an abundance of the materials we need, rather than accepting scarcity? It is possible, but that world cannot exist until we embrace biology as the transformative technology it truly can be.
Biology lets us build things at the atomic level, where changing what we want to make is a process of updating the information encoded in DNA. Biology is inherently open-source, available to everyone, and runs at room temperature using non-toxic materials. This isn’t science fiction: these are powers we already have, and biology is currently used to make food, medicines, chemicals, and building materials and already provides €2.4 trillion of Europe’s economy.
Much of the world doesn’t recognise that biology isn’t just the study of nature, but one of the most important technologies in our hands today. Almost all the foods we eat are made using plants and animals that are the result of centuries of selective breeding.
Similarly, many of the most advanced medicines are made using engineered bacteria to manufacture them, by brewing them in large tanks similar to how we brew beer or wine. Thanks to recent advances in DNA sequencing and editing, we have entered an era where we have begun to rationally engineer how living things work, and while these technologies are still in their infancy, products from the direct engineering of biology are already making over $350 billion per year in the United States.
A biological revolution
Fundamentally, this ability to change a living thing, such as a micro-organism, and then observe the effect, is revolutionising how we understand biological systems, leading to a full-scale transformation of biology from a science to an engineering practice. Understanding and ultimately engineering biology could make it much faster to develop new drugs such as antibiotics, to more efficiently produce the foods and materials we need to raise global standards of living, and to help clean up the toxic byproducts that have polluted the environment as a result of previous industrial revolutions.
With this focus on using the power of modern genomics to engineer biology, however, has come an awareness of the many limitations in how we presently think about and work with biology. This starts with the lack of reproducibility in the basic working practices we use in laboratories worldwide to conduct the experiments that try to unravel the complexities of biological systems.
When the results of an experiment vary by who conducts it, whether they are involved in the development of a micro-organism used to make an industrial chemical, or in understanding how a human cell line can become cancerous, then the conclusions we can draw are also limited.
The trouble with ‘shake vigorously’
At the heart of this reproducibility crisis is that we continue to design, conduct, interpret, and share the results from our biological experiments manually. We take for granted that information technology allows us to share a document instantly with an email, and make a physical copy using a printer from any hardware vendor. Labs do not share a similar digital language for describing how they work, tracking what they actually do, or translating an experiment from the hardware in one lab to another. Without a common unambiguous language for how work in biology is performed, variable interpretations of vague statements like “shake vigorously” introduce major differences in how scientists will perform an experiment.
Even worse, working primarily by hand limits the scope of an experiment we can physically conduct, as we are limited by the amount of complexity we can handle, both with our brains and our hands. As a result, biology is filled with small experiments that cannot provide definitive insight into these complex systems. At the same time, easy-to-collect data, such as genome sequences, is gathered on an industrial scale, rather than the difficult-to-collect data such as the effects that changes in these genomes create. Powerful new advances in machine learning, like deep learning, are therefore bottlenecked without having sufficient quantities of structured data to learn from.
What can biology learn from the iPhone?
One of the best examples of how tools help us to understand complex problems has been the semiconductor industry, which has famously doubled the number of transistors that can be designed and manufactured in a device every 18 months. This has happened while the number of chip designers has remained the same, because designers have become exponentially more productive through the power of digital languages, such as VHDL and Verilog. Having a digital representation of their problem, and tools to help rationalise, simulate, and ultimately automate the manufacture of chip designs written in these languages, has been the key reason iPhones are now packing more power than a supercomputer of decades past.
Empowering our work with biology with a similar foundation is absolutely vital to allow biology to fulfill its true potential. Just like there is no way to design a modern i7 processor from Intel by hand, there is no way to design and understand an entire genome with a piece of paper.
A digital representation of how we work in the lab is required for reproducible and scalable work. If we can finally build on one another’s work, we will gain a fuller understanding of biology’s potential for solving our complex global challenges.