Seedlabs

TCR-Mimic Foundry: De Novo scFv Design Against Peptide-MHC Complexes

A service that creates antibodies to target specific proteins on the surface of tumor or virus-infected cells.

A design service that produces antibody fragments binding a chosen peptide-MHC complex on the cell surface, turning intracellular tumor or viral antigens into antibody targets — demonstrated de novo on a PHOX2B peptide-MHC complex.

MedicineMonoclonal and Polyclonal Antibodies Research
Microbial Symbiosis and ResistanceImmuno-oncology (TCR-mimic antibodies and bispecifics targeting intracellular tumor antigens presented on HLA)

Concept

A design service for immuno-oncology teams that want to reach intracellular antigens. The customer names a disease-relevant peptide (tumor antigen, neoantigen, or viral epitope) and the HLA allele it is presented on; the service models or takes the peptide-MHC structure, designs heavy- and light-chain CDRs de novo against the peptide-HLA interface, screens by yeast display, and delivers scFvs convertible to full IgG or bispecific formats. The wedge: TCR-mimic antibodies let standard antibody technology target intracellular proteins once a fragment of them is displayed on MHC class I — but pMHC complexes are weak immunogens, and conventional discovery (immunization, patient isolation, random libraries) rarely yields them. De novo computational design sidesteps the immunization bottleneck, and loop-level precision matters because the peptide contributes only a small patch of the pMHC surface, so the binder must be geometrically exact to distinguish the target peptide-HLA from the vast background of self-HLA.

Why now

Paper [0] reports de novo scFv design against a PHOX2B peptide-MHC complex by combining designed heavy-chain and light-chain CDRs — evidence that fully computational design can address the pMHC target class that conventional antibody discovery struggles with. The companion TcdB scFv cryo-EM structure verified atomically accurate conformations of all six CDR loops (backbone RMSDs 0.2–1.1 Å), demonstrating the design method's control over loop geometry. Conversion of a designed scFv to a full-length IgG1 was also demonstrated (Kd 68 nM), providing a path from a designed fragment to a therapeutic-format molecule.

Written by GLM · 9 Oct 2026

AI assessment

Backed by 1 paper86

A high-value specialized design service that leverages RFdiffusion to solve the specific, difficult problem of targeting pMHC complexes for immuno-oncology.

Evidence strength
5/5
The full text provides explicit evidence of atomically accurate scFv design against a PHOX2B pMHC complex and successful conversion to IgG1, directly supporting the core technical claim.
Market pull
4/5
There is strong, existing demand from specialized biotech firms (e.g., Immatics) for TCR-mimics to target intracellular antigens, which are otherwise 'undruggable' by antibodies.
Novelty & moat
3/5
While similar to EpitopeLock, this focuses on the specific pMHC geometry and the high-stakes TCR-mimic niche, which requires higher precision than general epitope targeting.
Feasibility
4/5
The pipeline (RFdiffusion -> Yeast Display -> Characterization) is already demonstrated in the paper, making a prototype service highly feasible for a skilled team.
Wedge clarity
5/5
The wedge is exceptionally sharp: providing binders for specific peptide-HLA combinations that are notoriously difficult to discover via traditional immunization.
Simplicity / focus
5/5
The idea is a single, focused design service with one clear input (peptide/HLA) and one clear output (scFv/IgG).

Scored by Gemma 4 · 9 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Who benefits

  • Eureka Therapeuticscompany

    Built its pipeline on TCR-mimic antibodies; de novo pMHC design could expand the set of addressable peptide-HLA targets beyond what immunization-based discovery yields.

  • Immaticscompany

    Identifies tumor antigens presented on HLA and develops TCR-based therapeutics; pMHC-specific antibody binders would accelerate target validation and enable bispecific formats.

  • Medigene AGcompany

    Develops TCR-based immunotherapies; pMHC-binding antibody fragments support target validation and engineered-receptor programs against the same presented antigens.

  • Miltenyi Bioteccompany

    Research reagent and cell-analysis company; anti-pMHC antibodies are scarce, high-value reagents for detecting antigen presentation on tumor cells.

  • St. Jude Children's Research Hospitalorganization

    PHOX2B is a defining neuroblastoma driver; the paper's PHOX2B pMHC binders are directly relevant research tools for its pediatric solid tumor programs.

Research it builds on

  1. Atomically accurate de novo design of antibodies with RFdiffusion
    Nathaniel R. Bennett, Joseph L. Watson, Robert J. Ragotte et al. · 2025 · 305 citations
    All ideas from this paper →
Leave feedback
feasibility