TVs and radios blare that “artificial intelligence is coming,” and it will take your job and beat you at chess.
But AI is already here, and it can beat you — and the world’s best — at chess. In 2012, it was also used by Google to identify cats in YouTube videos. Today, it’s the reason Teslas have Autopilot and Netflix and Spotify seem to “read your mind.” Now, AI is changing the field of synthetic biology and how we engineer biology. It’s helping engineers design new ways to design genetic circuits — and it could leave a remarkable impact on the future of humanity through the huge investment it has been receiving ($12.3b in the last 10 years) and the markets it is disrupting.
The idea of artificial intelligence is relatively straightforward — it is the programming of machines with reasoning, learning, and decision-making behaviors. Some AI algorithms (which are just a set of rules that a computer follows) are so good at these tasks that they can easily outperform human experts.
Most of what we hear about artificial intelligence refers to machine learning, a subclass of AI algorithms that extrapolate patterns from data and then use that analysis to make predictions. The more data these algorithms collect, the more accurate their predictions become. Deep learning is a more powerful subcategory of machine learning, where a high number of computational layers called neural networks (inspired by the structure of the brain) operate in tandem to increase processing depth, facilitating technologies like advanced facial recognition (including FaceID on your iPhone).
[For a more detailed explanation of artificial intelligence and its various subcategories, check out this article and its flowchart.]
Regardless of the type of AI, or its application, we are in the midst of a computational revolution that is extending its tendrils beyond the “computer world.” Soon, AI will impact the medicines you take, the fuels you burn, and even the detergents that you use to wash your clothes.
Biology, in particular, is one of the most promising beneficiaries of artificial intelligence. From investigating genetic mutations that contribute to obesity to examining pathology samples for cancerous cells, biology produces an inordinate amount of complex, convoluted data. But the information contained within these datasets often offers valuable insights that could be used to improve our health.
In the field of synthetic biology, where engineers seek to “rewire” living organisms and program them with new functions, many scientists are harnessing AI to design more effective experiments, analyze their data, and use it to create groundbreaking therapeutics. Here are five companies that are integrating machine learning with synthetic biology to pave the way for better science and better engineering.
- Riffyn Catalyzing clean data collection and analysis
(Oakland, CA, founded in 2014, has raised $24.9M)
Machine learning algorithms must begin with large amounts of data — but, in biology, good data is incredibly challenging to produce because experiments are time-consuming, tedious and hard to replicate. Fortunately, one company is addressing this bottleneck by making it easier for scientists to do exactly that.
Riffyn’s cloud-based software platform helps researchers standardize, define, and perform experiments and streamlines data analysis, which enables researchers to focus on doing the actual science and makes the use machine learning algorithms to extract deeper insights from their experiments an everyday reality.
With this platform, experiments can be conducted more efficiently, leading to massive decreases in cost, improvements in productivity and quality, and data that is primed to be further analyzed with sophisticated machine learning techniques. That means companies can use this technology to develop new proteins for cancer therapeutics, and they can do it much faster and better than before. Riffyn already works with 8 of the top 15 global biotech and biopharma firms — and they were founded just five years ago.
- Microsoft Research Station B: Putting together the puzzle pieces of programming biology
(Cambridge, UK, officially launched in 2019)
There are a lot of moving parts in the synthetic biology world, which makes it difficult but vital to streamline and integrate operations as much as possible. For the last decade, the computational biology arm of Microsoft Research, Station B, has been developing machine learning models for biology to fix this problem and expedite research across a variety of fields, from medicine to construction.
Its efforts are paying off in the form of various new partnerships, too. With Synthace, it is developing software to automate and expedite experiments in the lab. Station B is additionally working with Princeton to research the mechanisms behind biofilms (relevant to how bacterial colonies develop antibiotic resistance) by utilizing machine learning-based methods that extract patterns from images taken during different stages of bacterial growth. Station B is also collaborating with Oxford Biomedica, a company harnessing these machine learning capabilities to improve a promising gene therapy for leukemia and lymphoma. This is perhaps one of synthetic biology’s biggest areas for impact: designing therapeutics to combat a variety of diseases.
- Atomwise: Deep learning decoding the black box of structural protein design
(Based in San Francisco, CA, founded in 2012, has raised $51M)
Atomwise is tackling drug development with their deep-learning platform, called AtomNet, that can rapidly model molecular structures. It can accurately analyze chemical interactions within small molecules to predict the efficacy of targeting diseases ranging from Ebola to multiple sclerosis. By utilizing data about atomic structure, Atomwise designs novel therapeutics that would otherwise be nearly impossible to develop.
They have numerous academic and corporate partnerships with institutions including Charles River Laboratories, Merck, University of Toronto, and Duke University School of Medicine, that are providing many of the real-world applications and opportunities to drive this research forward. They also recently announced an up-to $1.5B collaboration with the Jiangsu Hansoh Pharmaceutical Group, the Chinese company with one of this year’s biggest biopharma IPOs.
While Atomwise’s approach to designing molecules is powerful and well on its way to combatting multiple diseases, there is no one perfect method to computational discovery. That’s where Arzeda comes in.
- Arzeda: Rewriting the rules of protein design with de novo deep learning
(Seattle, WA, founded in 2008, has raised $15.2M)
Arzeda, a company originating from the Baker Lab at the University of Washington, uses its protein design platform (rooted, of course, in machine learning algorithms) to engineer proteins for everything from industrial enzymes to crops and their microbiomes.
Arzeda builds its molecules entirely from scratch (or de novo), rather than optimize existing ones, to perform new functions not found anywhere in nature; deep learning techniques are vital to ensure the proteins they design fold correctly (a very computationally demanding problem) and function as intended. Once the computational steps are complete, the new proteins are produced through fermentation (just like beer), bypassing natural evolution to efficiently produce brand-new molecules.
- Distributed Bio: Revolutionizing the future of the flu, cancer, snake bites, and more
(South San Francisco, CA, founded in 2012, self-funded by licensing technologies)
On the other end of the design spectrum, Distributed Bio harnesses rational protein engineering to optimize existing antibodies, which are the proteins in your body that detect bacteria and fight off other disease-causing invaders, to create novel therapeutics.
Among the many immunology-engineering technologies that the company boasts (from a universal flu vaccine to a broad-coverage snake antivenom) is the Tumbler platform. Using machine learning methods, Tumbler creates over 500 million variations of a starting antibody to expand and quantify the search space of what changes to the molecule are most valuable; then, it scores sequences to predict how well they bind to their target in real life and uses the “valuable change” information to further improve the best-scoring sequences. The production cycle continues as the top sequences are synthesized and tested in the lab. Eventually, an archetypal molecule emerges to fulfill the intended therapeutic purpose — something not necessarily observed in nature, but combining all of the best possible characteristics.
Tumbler has helped to enable a wide range of applications beyond traditional single-target drug development — from designing antibodies that bind to multiple targets simultaneously to creating chimeric antigen receptor T-cell (CAR-T) therapies (together with Chimera Bioengineering) for cancer treatments with reduced toxicity, the power of this end-to-end optimization platform to generate ideal antibodies at scale is unprecedented.
While this progress is exciting, artificial intelligence is not a universal replacement for our investigations of the natural world, nor is it the only way to develop cures for human diseases. At times, it may not be technically useful or even ethically sound. As we continue to reap the benefits of this technology and increasingly incorporate it into our daily lives, we must continue having conversations about the design, implementation, and ethics of innovations in synthetic biology and AI; we stand on the precipice of a new age for science and humanity.
Thanks to Aishani Aatresh for additional research and reporting in this article. Aishani is also a researcher at Distributed Bio developing computational immunoengineering methods to generate superior antibodies. Please note: I am the founder of SynBioBeta, the innovation network for the synthetic biology industry, and some of the companies that I write about are sponsors of the SynBioBeta conference (click here for a full list of sponsors).
Originally published on Forbes https://www.forbes.com/sites/johncumbers/2019/09/16/meet-5-synthetic-biology-companies-using-ai-to-engineer-biology/0