Astromech Raises $20M to Build Biological Operating System That Forecasts Evolutionary Change

Astromech raises $20M to build a biological operating system that can forecast evolutionary change

Astromech raises $20M to build a biological operating system that can forecast evolutionary change An artificial intelligence startup called Astromech that's developing models that can predict biological change has raised $20 million in funding, lifting its valuation to $3.8 billion. The round was led by biotechnology investor Bob Nelsen and saw the participation of Peak 6, NeoGenesis Capital, Builders VC and CAZ Investments, bringing the company's total amount raised to date to $60 million. Astromech was co-founded by the technology entrepreneur Ben Lamm (pictured) and the geneticist George Church in order to develop what they call a "predictive model of biology." The model crunches a combination of genomic, evolutionary, biological and functional data to try and anticipate how living systems will change over time, identify where they are vulnerable, and understand what's driving those changes. Whereas other biotech firms have tried to analyze organisms and diseases as they exist now, Astromech wants to forecast their future biological trajectories. The idea is that by doing this, it might be able to anticipate genetic bottlenecks, disease progression, drug resistance and how different species will respond to environmental changes. Lamm told Inc.com in an interview that the company's approach is not so different from trying to forecast the weather. Meteorologists, he explained, consider both historical weather patterns and current conditions to try and work out what will happen next. "We are building an algorithmic prediction solution," he explained. "Think of it like the weather, a complex system that humanity can predict due to specific technology and datasets. We are building the same thing for biology with evolutionary data." Astromech was spun out of the genetic engineering firm Colossal Biosciences Inc., which is trying to de-extinct animals such as the woolly mammoth, the Tasmanian tiger, the dire wolf, the dodo, the moa and the bluebuck. Inspired by Colossal's work, it leverages 3.8 billion years of evolutionary history as one of its primary training signals. Its model architecture blends genomic data from living and extinct organisms with evolutionary ancestry and biological responses. It has developed two separate model engines that work together to crunch all of this data - one that uses deep learning to spot patterns across different species, and another that's used to reverse-engineer biological systems back through history before projecting them forward to try and understand how they'll change in future. According to co-founder George Church, the startup distinguishes itself from others through its focus on ancestral regulatory states, rather than just proteins. He said that when we compare living species by their genomes alone, this method fails to assign DNA differences to functional impacts. "Most of the variations that matter for complex traits, for example, morphology and longevity are regulatory rather than coding, so reconstructing the ancestral regulatory state, not just the ancestral protein, has crucial explanatory power," he said. The startup is especially interested in the possibility of extending human longevity, and has chosen this area as a proving ground for its technology. So far, it has mapped 46 longevity-associated genes across a time-calibrated "tree of life" in order to examine how genes associated with cellular maintenance and cancer resistance have evolved. The goal is to try and identify the specific genomic and regulatory mechanisms associated with aging, which it believes could be the key to extending human lifespans. Lamm points to species such as the Asian elephant, which has evolved a unique cancer-suppression mechanism, and the bowhead whale, which typically lives for more than 200 years. By understanding how these animals have evolved, Astromech hopes to come up with extremely precise hypotheses that can aid in future longevity research. "When you can look across species and billions of years of evolution to understand why some biological systems are more resilient than others, you start asking really interesting questions about human health and longevity," he said. The immediate goal for Astromech is to expand its research teams, scale its comparative genomic infrastructure and study more species in order to get its hands on more evolutionary data. Once that's done, it aims to launch a number of pilot projects with partners in the health and biosecurity industries, where it will try and apply its vulnerability forecasting models to real-world challenges.

This AI Company Is Using 3.8 Billion Years of Evolution to Predict the Future

Ben Lamm has made a habit of building companies around ideas that sound a little crazy when you first hear them. Lamm, the serial entrepreneur behind Colossal Biosciences, has teamed up again with geneticist Dr. George Church to build Astromech, an AI company attempting to turn 3.8 billion years of evolution into something businesses have long had for weather, markets and consumer behavior: a forecasting system. Investors clearly see potential. Astromech, the latest company from Lamm and world-renowned geneticist Dr. George Church, has raised a new round of funding that values the company at $3.8 billion -- a fitting and ironic milestone for a startup building AI models informed by 3.8 billion years of evolutionary history. But what makes Astromech interesting isn't simply the valuation. It's the way Lamm is thinking about the business opportunity surrounding biology. "We are building an algorithmic prediction solution," Lamm shared. "Think of it like the weather -- a complex system that humanity can predict due to specific technology and data sets. We are building the same for biology with evolutionary data." Turning evolution into a prediction engine The concept is ambitious. Astromech combines genomic, evolutionary, and functional data to study how living systems have changed over time. Its AI models can then use those patterns to forecast how a genome, pathogen, or population could change next, where vulnerabilities might emerge, and what biological mechanisms are driving those changes. In other words, Lamm isn't building AI simply to analyze biology. He wants to use it to anticipate biology. That distinction could create opportunities across pharmaceuticals, drug discovery, pandemic preparedness, and synthetic biology. Rather than waiting for a biological problem to emerge and then reacting, companies could potentially identify risks earlier.