Machine learning platform screens ligands targeting CB1 and CB2

CB-Smart: A Cloud-Based Machine Learning Platform for Screening Cannabinoid Receptor Ligands.

ACS omega • • Highly Relevant
🤖

AI Summary

This study asked whether machine learning could help screen compounds for binding to the cannabinoid receptors CB1 and CB2, which are involved in processes including pain modulation and immune regulation. The researchers trained and compared several supervised machine-learning models using experimentally validated ligands and decoy compounds. The abstract does not report numerical performance results, so the size of the models’ predictive advantage cannot be assessed here.

Models using ECFP molecular fingerprints performed best among the tested approaches, and a hard-voting ensemble of the leading classifiers further improved predictive performance. The team also used SHAP analysis to identify molecular features associated with predicted binding and incorporated classification and affinity-prediction models into a cloud-based screening platform. This is an abstract-based summary: the abstract does not establish how well the platform performs in prospective experiments or whether its predictions lead to effective treatments.

💡 Key Findings

1
Among the tested models, those using ECFP molecular fingerprints—particularly XGBoost, SVM, and MLP—had the strongest reported predictive performance on reserved test datasets and an external dataset.
High
80%
2
A hard-voting ensemble combining the three leading classifiers further enhanced predictive performance, according to the abstract.
Good
75%
3
The researchers integrated binding classification and relative-affinity prediction models into a cloud-based platform for screening compounds against CB1 and CB2.
High
90%
4
SHAP analysis identified structural and physicochemical features associated with predicted receptor binding; the abstract does not specify those features.
Good
75%

📄 Original Abstract

Cannabinoid receptors CB1 and CB2 are class A G protein-coupled receptors (GPCRs) implicated in pain modulation, neuroinflammation, immune regulation, and metabolic processes, making them important therapeutic targets. Conventional structure-based virtual screening for cannabinoid ligands remains computationally intensive and limited in throughput. To overcome this challenge, this work presents CB-Smart, a cloud-enabled machine learning (ML) platform to predict binding of ligands and their relative binding affinity to CB1 and CB2. Using curated experimentally validated ligands and GPCR decoy datasets, we trained five supervised ML algorithms, XGBoost, Support Vector Machines (SVM), Random Forests, Multi-Layer Perceptron (MLP), and k-Nearest Neighbors (KNN). Three featurization strategies, ECFP molecular fingerprints, RDKit 210 molecular descriptors, and Lipinski descriptors, were systematically evaluated. Models built on ECFP fingerprint features, particularly XGBoost, SVM, and MLP, demonstrated the highest performance on both reserved test datasets and an external dataset. A hard-voting ensemble integrating the top three classifiers further enhanced predictive performance. Further, SHAP-based interpretability analysis highlighted key structural and physicochemical drivers of cannabinoid receptor binding, offering insights into ligand-receptor interactions and structure-activity relationships. Finally, we integrated classification and regression models into an interactive, user-friendly cloud platform, enabling broad accessibility for virtual screening, medicinal chemistry optimization and early-stage drug discovery efforts targeting cannabinoid receptors.

Explore More Research

Stay informed about the latest cannabis science.

Your stash, decoded.