New Publication Highlight
Caniceiro A, Amorim AMB, Rosário-Ferreira N, Moreira IS
GPCR-A17 MAAP: Mapping Modulators, Agonists, and Antagonists to Predict the Next Bioactive Target Journal of Cheminformatics, 2025, 17, 102
DOI: 10.1186/s13321-025-01050-z
Scientific context and contribution
G protein–coupled receptors (GPCRs) remain the most prolific class of drug targets, yet systematically identifying new bioactive ligands and targets across the GPCR landscape remains a major computational and translational challenge.
In this work, we introduce GPCR-A17 MAAP, a cheminformatics and AI-driven framework designed to map modulators, agonists, and antagonists in a unified bioactivity space, enabling the prediction of previously unexplored GPCR–ligand relationships.
Key highlights include:
-
Unified representation of GPCR ligands, capturing functional class (agonist, antagonist, modulator) rather than binary activity
-
Integration of molecular descriptors and machine-learning models to construct predictive bioactivity maps
-
Target-centric and ligand-centric analyses, enabling forward and reverse inference
-
Practical utility for hypothesis generation, supporting ligand repurposing and discovery of novel GPCR targets
This approach shifts GPCR modeling from isolated predictions to continuous bioactivity landscapes, better reflecting pharmacological reality.
Impact and relevance
Published in a leading open-access Q1 journal, this study provides a scalable and interpretable AI framework for GPCR drug discovery. GPCR-A17 MAAP is particularly well suited for early-stage hit discovery, polypharmacology analysis, and strategic target exploration, aligning closely with current trends in AI-enabled medicinal chemistry.
📎 Read the full article:
https://doi.org/10.1186/s13321-025-01050-z
Leave a Reply