← Back

Wire · founder news, decoded · opportunities

Verseon bets physics can take drug design beyond AI's training data

Published

22 July 2026

Topic

opportunities

Sectors

Digital Health

Geography

United States

Source

Read at drugdiscoverytrends.com

Verified

Fusion42 · 22 July 2026 · Fusion42 review

Verseon, a computational drug design firm founded in 2002, claims its physics-based platform can discover novel small-molecule drugs outside the chemical space explored by AI training data, including reversible anticoagulants with lower bleeding risk than current therapies like Eliquis and Xarelto.

This Wire brief tracks EternaFusion, and It sits within Fusion42's coverage of Digital Health. Wire is Fusion42's founder-focused intelligence feed: each story is connected to the funds and startups it names — every one with a live profile on Raise or Scout — so founders can follow the capital and the momentum behind the headline rather than just the headline itself. Wire analysis is one of the live surfaces Arthur reasons over.

The Wire takeaway

If you're building AI drug discovery, you're training on 150 years of human chemistry — a few grains of sand in a 10^33-compound universe. Verseon's physics-first approach finds drugs AI cannot see, and their anticoagulant just cleared European patents with dramatically lower bleeding than Eliquis or Xarelto. The commercial model has shifted: it's not about better search of known space, it's about designing in unknown space.

Mentioned in this story

Related on Wire

Topics

Digital Health · drug-discovery · computational-chemistry · anticoagulants · ai-vs-physics · novel-candidates