2024 MIT R&D Conference: Startup Exchange Lightning Talks - Concerto Biosciences

CHERI ACKERMAN: Thank you. My name is Cheri Ackerman. I'm one of the co-founders and the CEO at Concerto Biosciences, where we are making it possible to tap into one of our world's largest untapped natural resources. And that is the microbes that live in us and around us. Your body is covered in billions of microbes, and so is the surface of every object in the world. Our food, animals, plants, our buildings. So products that tap into this microbial world are the inevitable future of medicine, food, and consumer goods.

Creating these products will unlock an incredible amount of value. But only a small fraction of this has been accessed today. And the major roadblock that stands between us and capturing this value is knowledge. It is very difficult to predict which microbes will make the best or the most effective products. So whoever can reliably predict microbial performance can unlock these products, and then capture this economic real estate.

Well, if prediction is our problem, we live in the age of AI, let's build an AI model that can take in all of the possible different microbial products and predict which ones will be effective. This is a very reasonable idea. And so it begs the question, why has this not happened yet? And we would argue that a major deterrent to building these models is having the right data. Microbes live in very complex ecosystems. In the same way that human behavior is shaped by all of the humans around us, microbial behavior is shaped by the microbes that are around them.

Unfortunately, the technologies that we have today to study microbes don't actually teach us very much about how microbes interact. And this leaves us trying to build models with very incomplete information. What we need is measurements of how microbes perform in the context of the other microbes around them. But this turns into a massive combinatorial math problem. If you have 1,000 microbes and you want to measure all the different combinations, you're talking about millions and billions of combinations of microbes that you would have to construct and measure. And even the fastest robots today can't do that.

So while we were at MIT, we invented a technology called kChip. And kChip takes in nanoliter droplets of microbial cultures and randomly combines them into combinations. We can make millions of combinations, and then measure how each of those combinations performs using microscopy, like the image that you see here. With kChip we have breached the threshold of data sets that are large enough to train an AI model of microbial performance. The academic prototype of kChip was already producing data sets that were 10 times larger than the next largest data available in academia.

And since starting Concerto in 2020, we've gained another order of magnitude, making Concerto the sole owner of the data set that enables this kind of AI modeling of microbial performance. With this data, we've trained a model that we call kAI. And most importantly, kChip data, whether used alone or with kAI, enables us to find products that address high unmet need applications.

One of the areas that is ripe for disruption by microbe-based ingredients is skin and scalp care. The majority of ingredients in this space are decades old, leaving consumers frustrated and bored. Microbial ingredients, on the other hand, offer novel, gentle, and sophisticated approach to skin care, allowing brands to establish leadership with their consumers. One of the major microbes that causes these skin conditions is staph aureus. When the skin microbiome is disrupted, staph aureus becomes virulent. And it causes dryness, irritation and rashes.

However, the healthy skin microbiome normally pacifies staph aureus through these interactions with staph. So we can imagine designing an effective product to treat dryness, irritation and rash by finding combinations of microbes that inhibit staph aureus virulence. Using kChip, we made millions of combinations of skin microbes, measured how they influence staph aureus, and then use this massive data set to find a new therapeutic for atopic dermatitis. That's in a Phase I B study now. As well as in consumer product ingredients for skin irritation. And we're working on formulating those.

kChip can be applied to industries far beyond skin and scalp care. We've done successful pilot programs in agriculture, food ingredients, consumer care. We're actively seeking partners, both for our existing assets and to co discover new products, especially in human health, skin and scalp care, vaginal health, and food ingredients. So be very happy to talk about any of those areas with you at the break. Thank you.

[APPLAUSE]