A Dallas-based AI startup is building predictive models that can forecast how living systems will change, potentially helping scientists get ahead of genetic bottlenecks, disease progression, and drug resistance before they occur. Astromech, which spun out of Colossal Biosciences, closed a $20 million funding round and now sits at a $3.8 billion valuation, bringing its total raised to $60 million.

The company's core insight is deceptively simple: every genome carries a historical record of what changed, when it changed, and what biological tradeoffs followed. But almost none of that history is readable at scale today. Astromech was built to close that gap by training AI models on vast genomic datasets spanning extinct and living species, then using those patterns to anticipate where biological systems are headed.

How Does Evolutionary AI Actually Work?

Co-founded by Ben Lamm and Harvard geneticist George Church, Astromech compares its approach to weather forecasting. Just as meteorologists combine current conditions with historical patterns to predict future storms, the company uses functional genomic data and computational biology to trace how genes tied to longevity, cancer resistance, and cellular repair have stayed the same or changed across millions of years. The difference is scope: instead of testing one gene in one species, researchers can now trace these trends across entire groups of species simultaneously.

The company's first major demonstration showcased a map of 46 longevity-associated genes across a time-calibrated tree of life, a species family tree scaled to when each lineage split off. This allows scientists to see not just what genes exist, but how they've evolved and what that evolution tells us about biological vulnerability and resilience.

"Biology runs the world and historically, we have only reacted to it. We can describe biology in extraordinary detail, yet we still struggle to anticipate what comes next," said Ben Lamm, co-founder of Astromech.

Ben Lamm, Co-founder of Astromech

What Real-World Problems Could This Solve?

The practical applications span medicine and biosecurity. Asian elephants, for example, have evolved remarkable cancer-suppression mechanisms despite their size and long lifespans. Tasmanian devils, by contrast, remain vulnerable to a transmissible cancer that has devastated wild populations. Understanding why these divergent outcomes exist could unlock insights into human cancer resistance and inform therapeutic research.

Another striking example involves the TP53 gene, which plays a crucial role in cancer prevention. Elephants have 20 copies of this gene, while humans have only one. Astromech traces when that expansion happened across mammalian lineages and what the implications are for cancer incidence across species, including humans. This kind of evolutionary perspective could reveal why some animals resist cancer better than others and what genetic strategies might be worth exploring in human medicine.

With the new $20 million funding round, Astromech plans to expand its research team, increase the number of species represented in its functional genomic datasets, and scale the infrastructure used to train its models. The company is currently hiring for roles across ancestral modeling, regulatory genomics, genomic inference, sequence reconstruction, metabolic modeling, and protein folding.

Steps to Scaling Predictive Biology

• Expand Genomic Data: Increase the number of species represented in functional genomic datasets to improve model accuracy and breadth of predictions across the tree of life.
• Run Forecasting Pilots: Launch pilot programs with partners in health and biosecurity to test predictions in real-world scenarios and validate the model's ability to anticipate biological change.
• Build Early-Warning Systems: Convert predictive insights into actionable early-warning systems that can alert researchers to emerging vulnerabilities in biological systems before they become critical problems.
• Develop Therapeutic Programs: Use evolutionary insights to inform new therapeutic research programs that target the genetic mechanisms underlying disease resistance and longevity.

The funding round was led by Bob Nelsen, co-founder and managing director of ARCH Venture Partners, with participation from PEAK6, NeoGenesis Capital, Builders VC, and CAZ Investments. Nelsen was already an investor in Colossal Biosciences, Astromech's parent company, signaling confidence in the spinoff's direction.

George Church, the Harvard geneticist and Astromech co-founder, emphasized that the company's approach depends on tools that didn't exist a decade ago. Most variations that matter for complex traits are regulatory rather than coding, meaning scientists need to reconstruct the ancestral regulatory state, not just the ancestral protein sequence. This requires functional data across many species rather than sequence data alone, combined with AI reconstruction cheap enough to run genome-wide.

"Most of the variations that matter for complex traits are regulatory rather than coding, so reconstructing the ancestral regulatory state, not just the ancestral protein, has the crucial explanatory power. That takes functional data across many species rather than sequence alone, and AI reconstruction cheap enough to run genome-wide," explained George Church, co-founder of Astromech and Harvard geneticist.

George Church, Co-founder of Astromech and Harvard Geneticist

Astromech's work builds on Colossal Biosciences' broader mission and resources. Colossal made headlines for its de-extinction efforts, becoming Texas's first "decacorn" startup with a valuation exceeding $10 billion in January 2025. The parent company announced the birth of the Colossal Woolly Mouse in March 2025, a significant step toward genome engineering for de-extinction, and hatched live chicks using a 3D-printed shell in May 2026. These achievements have generated both scientific credibility and investor confidence that now extends to Astromech's evolutionary forecasting platform.

Over time, Astromech plans to turn its work into a global platform that can anticipate biological change across diverse applications. The company describes itself as being in a deep research and development phase, with its platform and initial research pipelines already operational. The next phase includes running forecasting pilots with partners in health and biosecurity, marking a transition from foundational research to real-world validation.