Harnessing AI-driven reverse docking in drug discovery: a comprehensive review of opportunities, challenges, and emerging trends

Abstract

The integration of artificial intelligence (AI) with reverse docking methodologies is reshaping drug discovery by streamlining the identification of drug targets and therapeutic interactions. This approach is pivotal in drug repurposing, safety profiling, and predicting off-target effects. Reverse docking uniquely identifies potential binding sites across diverse protein targets, providing insights into drug efficacy and adverse outcomes. AI technologies, such as machine learning, deep learning, and reinforcement learning, enhance this workflow by optimizing target selection, virtual screening, and conformational sampling. Despite challenges like data limitations and algorithmic complexities, AI-driven reverse docking has shown promise in drug repurposing and precision medicine, as illustrated by successful case studies. This review highlights its transformative potential and future prospects, including the incorporation of multi-omics data and real-time discovery pipelines for personalized medicine.

The computational strategies discussed leverage reverse docking platforms integrated with AI frameworks. Machine learning and deep learning models were employed for target selection and interaction prediction, while reinforcement learning facilitated advanced sampling techniques. Virtual screening workflows incorporated AI-driven optimizations for docking simulations. These methodologies were implemented using widely recognized computational tools, including AI libraries and molecular docking software, ensuring robust and reproducible results. Challenges in data integration were addressed by employing high-throughput pipelines capable of processing multi-omics datasets, thus supporting comprehensive drug discovery initiatives.