RAEXENSE: Chemical Core of RECEPTOR.AI

REAXENSE serves as the core chemical resource for Receptor.AI ecosystem, bringing together predictive AI with real-world synthesis and validation.

REAXENSE dedicated chemical spaces

AI-Driven Chemical Space for High-Throughput Discovery

8M+ ready-to-use compounds, representing ~2.5M unique Murcko scaffolds, curated for hit discovery and lead optimization
90% Ro5-compliant molecules and high structural diversity (T_mean < 0.35)
All in-stock, validated purity ≥90%, available in vials, custom assay plates, and bulk formats for global delivery
Fully integrated in screening workflows, SAR modeling, and ADMET filtering using standardized data formats

AI-Navigable Virtual Space - 10¹⁶ Synthesis-Ready Molecules

10¹⁶ novel, diverse and synthetically feasible molecules derived from 1,000,000+ building blocks and 1,000+ reactions
Real-time synthetic feasibility (SAS<7) with on-demand synthesis
Optimized for drug-like and bRo5 compounds to maximize chemical diversity and therapeutic potential
Search 10⁹+ molecules in under a second using 256–2048-bit ECFP4/MHFP fingerprints; supports AI-driven hit discovery and IP-free design in novel chemical space

AI-Powered Platform for Therapeutic Peptide Discovery

Design 2–100 AA peptides—cyclic, branched, or modified—using 20 standard and 500+ non-standard AAs, with access to 10¹³ variants
AI-optimized for drug-likeness: supports stability (min–hrs), affinity (Kd: μM–nM), and real-time 3D modeling
SPPS-compatible up to 50 AAs with 95–99% efficiency and optional GMP formats
Enables sequence-to-function prediction, SAR optimization, and PPI-targeted design across oncology, immunotherapy, and infectious disease applications

AI-Driven Macrocyclic Discovery for Challenging Targets

De novo design of bridged rings and peptidomimetics (8–20+ atoms) using transformers, LSTMs, and GANs
Engineered for permeability and strong binding (Kd μM–nM) with TPSA ≤140Å², cLogP 2–6, MW 500–800 Da, and conformation-stabilizing H-bonds
Synthesis-ready with 80–95% per-step yields, >70% success in RCM, ~95% cyclization efficiency via NRPS
Integrated for ADMET prediction and PPI-targeted macrocycle design in PEGylated, lipidated, and nanoformulated formats

On-demand libraries for 8,801 fully characterized proteins with identified binding pockets and assessed druggability.

Fully characterized protein target (therapeutic relevance, PPIs, known ligands SAR)
Detected binding pockets and generated conformational ensembles
Integrated with ChemoVista^TM and VirtuSynthium^TM chemical spaces