Rayca Science Induced proximity
These medicines work by bringing two proteins together. Rayca designs the whole assembly, the target, the effector and the molecule between them, as one system.
Induced proximity medicines act through a complex that is rarely observed directly. Binary affinity alone leaves most of the biology unexamined.
Where R&D loses time and evidence today, and how Rayca addresses each point.
An unresolved ternary complex
Activity depends on a three-body assembly that can adopt many arrangements and is seldom determined experimentally. Affinity for either protein alone is a weak guide to efficacy.
How Rayca addresses it
The assembly as the unit of design
Candidate complexes are generated, followed with dynamics and scored, including how strongly the molecule favours each arrangement.
Empirical linker optimisation
Attachment points, linker length, rigidity and composition are commonly explored by synthesising series, at considerable cost per iteration.
How Rayca addresses it
Linkers designed for the geometry
Exit vectors and linkers are generated together and tested against the modelled complexes before synthesis.
Formation without function
A stable complex is not necessarily a productive one. Ubiquitination geometry, complex lifetime and the loss of effect at high concentration govern efficacy, and are rarely assessed before compounds are made.
How Rayca addresses it
Productive geometry assessed early
Reach, persistence, concentration response and selectivity are evaluated for each candidate, beyond affinity.
Developability beyond the rule of five
Bifunctional molecules occupy chemical space where conventional ADME models were not trained, so permeability, solubility and metabolic liabilities tend to emerge late.
How Rayca addresses it
Properties modelled for this space
Permeability, including folded conformations, solubility and stability are modelled directly and carried through every round.
The ternary complex 01
Design follows the assembly that drives the biology.
The assembly
- Target protein
- Effector complex
- Bridging molecule
Target
The protein the medicine is meant to act on.
Modelling the assembly 02
Cooperativity is estimated before synthesis, rather than discovered after.
Modelling the complex
- Sample arrangements
- Place the molecule
- Follow the dynamics
- Estimate cooperativity
Energy landscape
Candidate orientations of target and effector are generated.
Linkers and exit vectors 03
Linker choices follow the geometry the complex requires.
Designing the bridge
Where each ligand can be joined without losing its binding.
Permeability
Solubility
Metabolic stability
Size and flexibility
Productive geometry 04
Candidates are judged on the outcome they produce, beyond affinity.
Productive geometry
Reach
Accessible target residues near the ligase machinery.
Reach
Surface lysines of the target that sit inside the ligase’s working radius can receive ubiquitin. Those outside it cannot.
Beyond the rule of five 05
Developability is designed in from the first bifunctional candidate.
Properties of large molecules
Including folded shapes that shield polar groups.
Across modalities 06
One design system, with the right tools for each modality.
Methods by modality
Molecular glues
New protein interfaces, scored and designed.
- Interface scoring
- Surface complementarity
- Glue-site search
Small-molecule degraders
Ligands and linkers, designed together.
- Ligand and linker design
- Ternary modelling
- Property control
Biologic bifunctionals
Protein design with developability checks.
- Protein design
- Format and linker
- Developability