Adverse drug reactions (ADRs) are a major cause of clinical trial failure and post-market withdrawal, posing significant risks to public health and impeding drug development. While computational methods offer an alternative to costly preclinical testing, existing models often fail with novel compounds by requiring pre-existing information such as drug-ADR associations or by inadequately integrating diverse data sources. Here, we introduce DeepADR, a multi-modal deep learning framework for predicting both the occurrence and frequency of ADRs using early-stage, readily available data. DeepADR integrates chemical structures and biological target profiles with semantic representations of ADR terms derived from a large language model. These heterogeneous parameters are fused using a Kolmogorov-Arnold Network (KAN), which effectively models complex, non-linear cross-modal interactions to capture underlying toxicological mechanisms. Our model outperforms existing methods in predicting both ADR occurrence and frequency, demonstrating robust generalization to new chemical entities. By effectively integrating chemical, biological, and semantic datasets, DeepADR provides a powerful, scalable tool for the early-stage safety assessment and candidate prioritization. This framework not only facilitates the prioritization of safer drug candidates but also offers a methodology for predicting the toxicity of other hazardous materials, holding significant promise for advancing public health.
Wan et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: