Flagship Pioneering has committed $50 million to Montai Health, a newly launched company unlocking the power of nature with “anthromolecules”—molecules with a long history of safe, chronic human consumption—to enable earlier intervention and slow, preempt, and even halt disease progression.
The funding will support the development of Montai’s CONECTA platform and its initial pipeline of new anthromolecule medicines, which already has lead candidates in several different indications. The start-up is initially focusing on auto-immune diseases and building a comprehensive map across all inflammatory diseases.
“We start with the problems that we immediately need to solve, which is safer medications that have the potential, given their safety profile, to go in as early in the disease as possible,” Montai Health CEO and Flagship Pioneering CEO-Partner Margo Georgiadis told GEN Edge.
“There’s this huge pile of [medicinal] keys sitting in nature, and we’re using advances in chemical machine learning in the digitization of biology to find as many key-and-lock pathway combinations as possible. You can find much more interesting and complex molecules than people would normally pick out [in drug development today] that can have a biological modulating effect. The nature of these molecules will enable getting medicines to people more quickly”
Medicinal mountaineering with machine learning
At the heart of Montai Health is thinking about combinations of molecules meant to work together to treat and prevent the causes of disease, not just the symptoms. The startup, based in Cambridge, Massachusetts, has started gathering information about 100,000 molecules. The problem is trying to match each individual or combination of anthromolecule to biological pathways affected by disease.
That’s where Georgiadis comes in. She previously served as the CEO of Ancestry and President of the Americas at Google and wants to harness technology to improve lives in a transformative way. Named after getting to the top of a mountain with artificial intelligence (AI), Montai Health is creating a rapid and repeatable process to leverage insight from combinatorial models and other analytics to create these safe medicines for long-term use.
To complete the picture for each anthromolecule, Montai Health is leveraging the massive digitization of biology with in-house computational models that characterize anthromolecules on different elements essential for developing medicines, such as understanding and mapping the bioactivity of anthromolecules across all known biological pathways.
The next step in Montai Health’s computational process is to generate another set of models to search for the pathway solution. To do so, they create a bioactivity atlas for a disease by looking at all of the known pathways for that disease and then consider what an optimized pathway solution for that disease would be. The result of the computational pipeline is what Montai is calling “anthrographs,” which are models that take all the data about understanding each individual molecule—dosage information, drug interactions, and network information—to develop new medicines.
“By building this next-generation technology, it’s enabling us to unlock the richness to solve these problems and leverage combinations of [anthromolecules] to stay within safety profiles or have a greater inclusion profile because we can [target] multiple pathways,” said Georgiadis.
“Without AI/ML, that’s an infinite common combination space. If we didn’t have the ability to do this systematic testing and development, this would not have been possible even four or five years ago. If we can unearth this massive information systematically for as many diseases as possible, we will have the opportunity to have a huge difference in treatments for millions of people. These are huge diseases areas that have very poor solutions today.”
These models are then brought into the wet lab to ensure reproducibility in the predictions. “The beauty is in the blend; and for us, it’s the blend of deep understanding of the natural chemical space and how to systematically interrogate and build that anthrograph knowledge,” said Georgiadis. “Then you start to put these building blocks together and realize that’s what technology has always been about—aggregating and integrating all of these different data sets so you can fundamentally reimagine how you ask questions.”
Georgiadis thinks that the work at Montai Health could lead to personalized cocktails for people in the long run. “This is a solvable analytics problem,” said Georgiadis. “Of course, the human body is complicated, but when we start with a molecule that’s fundamentally safe, this becomes a lot more of a solvable problem. That’s what’s so exciting about what’s happening with the power of computational technologies.”
