COMET:Combined Matrix for Elucidating Targets

2024-12-03

AI Summary

To address the low accuracy and inefficiency in drug target identification, this study proposes DT-Fusion—the first synergistic matrix framework integrating chemical similarity, structure-based molecular docking, and machine learning–driven probabilistic modeling. DT-Fusion supports three core tasks: similar compound retrieval, 3D binding mode prediction, and probabilistic target ranking, with an average runtime of <10 minutes per task. It covers 2,685 human disease-related targets (including both experimentally validated and exploratory targets), leveraging molecular fingerprint comparison, deep learning–based feature extraction, a refined binding pocket library derived from PDBbind+, a lightweight docking engine, and an ensemble ranking model. On benchmark datasets (ChEMBL/BindingDB), DT-Fusion achieves an 80% Top-15 target hit rate—significantly outperforming five state-of-the-art methods. A free web service is publicly available.

Technology Category

• Machine Learning: Matrix & Tensor Methods
• Data Mining & Knowledge Management: Web
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Identifying the interaction targets of bioactive compounds is a foundational element for deciphering their pharmacological effects. Target prediction algorithms equip researchers with an effective tool to rapidly scope and explore potential targets. Here, we introduce the COMET, a multi-technological modular target prediction tool that provides comprehensive predictive insights, including similar active compounds, three-dimensional predicted binding modes, and probability scores, all within an average processing time of less than 10 minutes per task. With meticulously curated data, the COMET database encompasses 990,944 drug-target interaction pairs and 45,035 binding pockets, enabling predictions for 2,685 targets, which span confirmed and exploratory therapeutic targets for human diseases. In comparative testing using datasets from ChEMBL and BindingDB, COMET outperformed five other well-known algorithms, offering nearly an 80% probability of accurately identifying at least one true target within the top 15 predictions for a given compound. COMET also features a user-friendly web server, accessible freely at https://www.pdbbind-plus.org.cn/comet.

Problem

• Drug-Target Interaction
• Prediction Tool
• Drug Development

Innovation

• COMET
• Drug-Target Interaction Prediction
• High Accuracy