OnCo
technologiesTechnologyPhase 1

De novo designed protein binders

Designing a protein from scratch on a computer to grip a chosen target, instead of finding one in an animal or a library.

Deep-learning structure prediction and generative design now produce small, stable binders against chosen epitopes in weeks, including minibinders for viral and cancer targets. In oncology the first clinical examples are AI-designed antibodies rather than fully de novo scaffolds; Generate Biomedicines' MMAE-neutralising antibody entered the clinic in 2026. Designed miniproteins are attractive as ADC and radioligand targeting heads because they penetrate tissue faster than antibodies.

Generic schematic · not to scale · placeholder for the drug discovery front
Guide RNA library · Knockout → dependency

How it works

Diffusion and language models generate backbone geometries complementary to a target epitope; sequences are designed, expressed and screened, compressing discovery from years to weeks.

Strengths
  • Fast, cheap discovery cycles
  • Small size penetrates tumours and clears quickly, useful for imaging and radioligands
  • The epitope can be chosen deliberately
Limitations
  • Immunogenicity of non-natural sequences is uncharacterised in patients
  • Short half-life needs engineering
  • Clinical validation is only starting

Key papers

1top

Latest papers

top
Latest papers · live from Europe PMC
Open in Europe PMC

Query for this technology: (TITLE:"De novo designed protein binders" OR ABSTRACT:"De novo designed protein binders") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about De novo designed protein binders, not a curated reading list.

Connected

12top

Pages like this

not linked directly; found by shared links