PRISM BioLab and Receptor.AI collaborate on AI-guided drug discovery targeting protein-protein interactions

PRISM BioLab and Receptor.AI Enter Drug Discovery Collaboration to Target Protein-Protein Interactions with AI-Guided Platform

Tokyo-based PRISM BioLab Co., Ltd. and U.S.-based Receptor.AI Inc. have entered into a drug discovery collaboration aimed at building an integrated platform for identifying orally available small molecules against intracellular protein-protein interactions, membrane proteins, and complex receptor systems (announcement). The collaboration combines PRISM BioLab's proprietary PepMetics® chemistry with Receptor.AI's physics-informed generative AI engine. The initial therapeutic focus is metabolic disease, including obesity. Both companies are privately held.

Financial terms of the agreement were not disclosed. The announcement does not reference upfront payments, development or commercial milestones, royalties, equity components, or territorial rights. The deal appears structured as a research collaboration rather than a conventional licensing transaction with defined economic terms.

Deal context

PRISM BioLab contributes its PepMetics® platform, which produces conformationally rigid, three-dimensional small molecules designed to mimic alpha-helix and beta-turn peptide motifs. These structural features are commonly found at the interfaces of intracellular protein-protein interactions and receptor-ligand binding sites. The company maintains a virtual and physical library of these peptide-mimetic compounds. Receptor.AI contributes QuorumMap, a chemical space navigation engine that integrates structure-based and physics-informed modeling with multi-objective optimization. The AI system is designed to prioritize compounds that balance target binding with oral bioavailability, selectivity, permeability, and metabolic stability.

Under the agreement, Receptor.AI will perform molecular design using PRISM's chemical space combined with its own algorithms. The collaboration will initially focus on one undisclosed receptor target in metabolic disease. No specific target name, gene, or receptor has been identified publicly. No named drug candidates or molecules have been disclosed, and no clinical trials are associated with the partnership. The work is at the discovery stage.

The companies stated they intend to jointly approach pharmaceutical partners to license or co-develop programs generated through the combined platform. A stated objective is to acquire specialized know-how from the integration of both technologies, which would form the basis for future partnership proposals. PRISM BioLab's broader pipeline focuses on cancer, autoimmune disease, and fibrosis, though the collaboration's initial scope is limited to metabolic disease. Receptor.AI has disclosed prior collaborations with Ono Pharmaceutical and claims more than 40 completed joint discovery projects across its AI platforms.

Protein-protein interactions remain a difficult target class for conventional small molecule drug discovery. PPIs typically involve large, flat binding surfaces that lack the deep pockets exploited by traditional medicinal chemistry. Peptide-mimetic approaches attempt to address this by presenting three-dimensional scaffolds that can engage these surfaces while retaining the oral bioavailability and pharmacokinetic properties associated with small molecules. The application of AI-driven molecular design to this target class is an area of active interest across the industry, though few PPI-targeted small molecules have advanced to late-stage clinical development.

Several companies are pursuing AI-driven or structure-based approaches to PPI drug discovery or related difficult target classes:

• Fog Pharmaceuticals — alpha-helical peptide mimetics — preclinical to early clinical stage • Sapience Therapeutics — designed peptides targeting transcription factors — Phase II • Aileron Therapeutics — stapled peptides (ALRN-6924, targeting MDM2/MDMX) — clinical development paused

These programs vary in modality and are not direct comparators to the PRISM BioLab and Receptor.AI platform, but they reflect ongoing efforts to drug PPI targets using non-traditional small molecule or peptide-based strategies.

The collaboration does not include any disclosed clinical-stage assets. No trial records exist for molecules emerging from this partnership. The value of the deal will depend on whether the integrated platform can generate candidates with sufficient potency and drug-like properties to attract pharmaceutical partners or advance into preclinical development. The absence of disclosed financial terms and the early-stage nature of the work place this agreement in the category of exploratory research partnerships rather than asset-driven licensing transactions.

---

Spot something wrong? Report an issue with this article