ADGSyn: Dual-Stream Learning for Efficient Anticancer Drug Synergy Prediction

2025-05-25

AI Summary

To address the challenge of prohibitively large combinatorial drug space and infeasibility of exhaustive experimental screening in oncology, this paper proposes a dual-stream graph neural network framework for efficient and accurate prediction of drug synergy. Methodologically, it introduces a novel shared-projection and attention-driven cross-drug feature alignment mechanism; incorporates adaptive message passing (AMP) to reduce GPU memory consumption by 40% and accelerate inference threefold; and employs LayerNorm to stabilize residual gradient flow, enabling full-batch molecular graph training. Evaluated on the O’Neil dataset comprising 13,243 drug combinations, the model significantly outperforms eight state-of-the-art baselines. Notably, it supports full-batch training of up to 256 molecular graphs on a single GPU—setting a new efficiency benchmark for computational oncology.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Drug combinations play a critical role in cancer therapy by significantly enhancing treatment efficacy and overcoming drug resistance. However, the combinatorial space of possible drug pairs grows exponentially, making experimental screening highly impractical. Therefore, developing efficient computational methods to predict promising drug combinations and guide experimental validation is of paramount importance. In this work, we propose ADGSyn, an innovative method for predicting drug synergy. The key components of our approach include: (1) shared projection matrices combined with attention mechanisms to enable cross-drug feature alignment; (2) automatic mixed precision (AMP)-optimized graph operations that reduce memory consumption by 40% while accelerating training speed threefold; and (3) residual pathways stabilized by LayerNorm to ensure stable gradient propagation during training. Evaluated on the O'Neil dataset containing 13,243 drug--cell line combinations, ADGSyn demonstrates superior performance over eight baseline methods. Moreover, the framework supports full-batch processing of up to 256 molecular graphs on a single GPU, setting a new standard for efficiency in drug synergy prediction within the field of computational oncology.

Problem

• Predicting anticancer drug synergy efficiently
• Reducing computational cost in drug combination screening
• Enhancing accuracy in drug synergy prediction

Innovation

• Cross-drug feature alignment with attention mechanisms
• AMP-optimized graph operations for efficiency
• LayerNorm-stabilized residual pathways for stability