Google DeepMind has published research in Nature demonstrating that digital watermarks can be embedded into AI-designed proteins while preserving their biological functionality. The study, titled Function-preserving watermarking of AI-generated proteins, involved collaboration with Adaptyv Bio and experimental validation using Carterra's LSA-XT high-throughput surface plasmon resonance (HT-SPR) platform.
The research introduces SynthIDBio, a family of methods designed to incorporate identifiable provenance signatures into AI-generated protein sequences and predicted molecular structures. The approach aims to support traceability, scientific integrity, and biosecurity as artificial intelligence becomes increasingly integrated into protein engineering.
To evaluate whether watermarking affects protein performance, Adaptyv Bio conducted experimental binding assessments across hundreds of AI-designed proteins. Researchers examined binding kinetics and affinities against three therapeutically relevant targets: the SARS-CoV-2 receptor binding domain, VEGF-A, and PD-L1.
The study also introduces SynthIDBio-structure, a modified version of AlphaFold 3 that embeds watermarks into predicted biomolecular structures while maintaining comparable prediction accuracy.
Experimental validation was conducted using Carterra's LSA-XT platform, which combines microfluidics with real-time HT-SPR technology to measure molecular interactions. According to Carterra, its platforms can deliver up to 100 times the throughput of conventional label-free technologies while requiring smaller sample volumes.
“Adaptyv's role is to turn AI-designed proteins into real experimental data, quickly and at scale,” said Julian Englert, CEO and co-founder of Adaptyv Bio. “We use Carterra's HT-SPR for campaigns like this across hundreds of designs. It gives us real binding kinetics and affinity data, not just a yes/no answer — which is what lets the AI model learn and improve the next round of designs.”
The research reflects growing adoption of laboratory-in-the-loop workflows, where AI-generated molecular designs undergo experimental testing, with resulting data informing subsequent model development.
Carterra has also expanded its technology portfolio with the February 2026 introduction of Carterra Vega, a 48-channel HT-SPR platform capable of screening more than 20,000 small-molecule interactions daily.
Additionally, Carterra recently announced a collaboration with AstraZeneca focused on developing hardware and software for laboratory-integrated drug discovery workflows.
Click here for the original news story.
