2026-10-10T11:31:29Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34028306 2026-09-30T00:06:38Z category_13 category_873 category_734 category_931 category_69 category_106 portal_63 item_type_3 month_year_09_2026

Language of Toxicity: An eXplainable Artificial Intelligence Approach Nicola Amoroso (4591525) Ester Pantaleo (12400655) Fulvio Ciriaco (5569355) Francesca Cutropia (24130888) Nicola Gambacorta (9514363) Fabrizio Mastrolorito (14264646) Alfonso Monaco (4739133) Angelica Orfino (17699096) Roberto Bellotti (4591528) Orazio Nicolotti (168458) Genetics Chemical Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified Inorganic Chemistry Science Policy small molecules represents relevant molecular substructures predefined molecular descriptors often strongly imbalanced gated recurrent unit chemical safety assessment two important challenges based framework offers architecture also exploits drug repurposing applications related structural patterns model proposed combines capture chemical patterns unified architecture two languages proposed attention inspired framework drug development drug design sequential nature roc curve repeated cross potentially limiting nontoxic chemicals model provided future investigations future improvement fundamental challenge different scales capture complex canonical smiles average area attention mechanism accurate description Toxicity prediction in small molecules represents a fundamental challenge in drug development and chemical safety assessment. Traditional approaches heavily rely on predefined molecular descriptors or fingerprints, potentially limiting the ability to capture complex and nonlinear structure–activity relationships. Here, we present a descriptor-free, language-inspired framework that can be applied to different toxicity prediction tasks within a unified architecture. The model proposed combines a multiscale Convolutional Neural Network (CNN) layer to capture chemical patterns at different scales and a Gated Recurrent Unit (GRU) layer to capture the sequential nature of these patterns. This architecture also exploits an attention mechanism that computes attention weights across the sequence, enabling the model to focus on the most relevant molecular substructures for toxicity prediction. Toxic and nontoxic chemicals, represented by canonical SMILES, are investigated as the words of two languages which have to be discriminated; using eight different end points, the model provided an accurate description of toxicity patterns, with an average Area under the ROC curve (AUC) of 0.83 (min: 0.70, max: 0.94) under repeated cross-validation. The models were trained on relatively small data sets (∼1000 samples) and often strongly imbalanced, two important challenges that highlight the opportunities for future improvement; moreover, the proposed attention-based framework offers a representation of the molecular regions influencing model predictions, providing a basis for future investigations into toxicity-related structural patterns and potentially supporting hypothesis generation in drug design or drug repurposing applications. 2026-09-29T00:00:00Z Dataset Dataset 10.1021/acs.jcim.5c03214.s003 https://figshare.com/articles/dataset/Language_of_Toxicity_An_eXplainable_Artificial_Intelligence_Approach/34028306 CC BY-NC 4.0