Formula-Y designs antibodies in silico and validates them with partners. The current project turns that approach into a working proof of concept: an AI-driven design pipeline that returns a ranked shortlist of antibody candidates against a chosen target, ready for wet-lab validation.
What the project sets out to do
Antibody discovery is still largely driven by laboratory trial-and-error: screening large numbers of candidates and hoping the right one is among them. This project builds the alternative. Candidates are designed computationally, assessed before any experiment runs, and only the strongest go forward to the bench. The aim is a validated pipeline that forms the technical foundation of Formula-Y's antibody design service.
Where it stands
The project is underway. It is experimental and iterative by nature: we have been building and testing early versions of the pipeline, learning from each round and refining the approach. Top-ranked candidates will be handed off for in-vitro validation with laboratory partners, and results will feed back to sharpen the next round of designs.
Why it matters
By replacing rounds of blind laboratory screening with data-guided design, the approach reduces trial-and-error, lowers cost, and cuts the reliance on animal use in the discovery phase. It also anchors high-value, digital R&D in Groningen and the Northern Netherlands, in line with the region's analogue-to-digital innovation priorities that the VIA and Just Transition Fund programme supports.
Partner with us
We are looking for laboratories that can take AI-designed antibodies into in-vitro validation, and for research and biotech teams who want to build antibody programs together. If that sounds like you, we would love to talk.