Skip to content
#explainable ai Open access

From QSAR to deep learning: an interpretable comprehensive pipeline with a read-across approach for mutagenicity prediction via the Enalos Cloud Platform

Sep 2026 · RSC Advances · Vol 16, pp. 53776 - 53805 · 0 citations · 129 references
Medicine

TL;DR

In silico NAMs, including computational approaches, can contribute to the development of novel Safe and Sustainable by Design chemicals and substances by identifying potentially hazardous ones at an early stage by identifying potentially hazardous ones at an early stage.

Abstract

Assessing the mutagenicity of chemical compounds is essential for ensuring their safe handling and use, thereby minimizing potential health risks. New approach methodologies (NAMs), including computational approaches, provide non-animal alternatives for testing novel materials and chemicals. This study highlights the potential of in silico NAMs, which can contribute to the development of novel Safe and Sustainable by Design (SSbD) chemicals and substances by identifying potentially hazardous ones at an early stage. Emphasis is given to the mutagenicity prediction based on data from the Ames (bacterial gene mutation) test curating them to consider stereo-specific input whenever necessary. A consensus strategy integrating different chemical representations (molecular fingerprints, 2D and 3D descriptors and molecular graphs) and modelling methods, i.e., Quantitative Structure–Activity Relationship (QSAR), read-across and deep learning models, is employed to predict the mutagenic profile of chemical compounds. In this course, a XGBoost model is developed based on molecular descriptors and a graph convolutional neural networks model to classify compounds as mutagens and non-mutagens based on the Ames test data. The devised read-across methodology is based on a guided-k-Nearest Neighbours scheme (guided-kNN) where two different molecular representations (molecular fingerprints and descriptors) are considered for neighbour selection and predictions generation. The mutagenicity predictions from the three models are integrated in a majority voting scheme to enhance the overall predictive accuracy (83% in external validation) and reduce individual model biases. Interpretation of the descriptors involved in prediction is performed through explainable AI (XAI) methods to provide insight to the mutagenicity mechanism. To enhance the interpretability of the XAI-derived insights and reinforce user confidence in the models' predictions, the involved descriptors are mapped to key events leading to mutations within the Adverse Outcome Pathway (AOP) networks. Apart from the development of reliable and interpretable mutagenicity models, emphasis is given on delivering a pipeline for the generation of 3D descriptors that can be used as the basis for future cheminformatics models. To support transparency and reproducibility of the results of our work, the curated mutagenicity dataset used for modelling is disseminated through the ChemPharos database (https://db.chempharos.eu/datasets/Datasets.zul?datasetID=ds18), the modelling steps are documented following the standardized Modelling Data (MODA) guidelines and the consensus model is freely available via the Enalos Cloud platform (https://www.enaloscloud.novamechanics.com/insight/polis/), to facilitate virtual screening of novel compounds.

Read PDF

Similar papers

Open access Aug 2026

ToxiVerse: chemical bioprofiling, toxicity data sharing and customizable predictive modeling

ToxiVerse provides automatic chemical bioprofiling, curated toxicity datasets, and a predictive modeling interface designed for researchers who lack programming expertise to address the urgent need for user-friendly machine learning tools in computational toxicology.

Prasannavenkatesh Durai, Daniel P. Russo, Yitao Shen et al. · 0 citations
Open access Sep 2026

From Chemical Scaffold to Predicted Toxicity: A Scaffold-Aware Pipeline for Reliable In Silico Prioritization of Heracleum Furanocoumarins

A reproducible computational pipeline that evaluates not the predictions themselves but their reliability of furanocoumarins of the genus Heracleum as a model class of natural products, indicating that the predicted toxicological profile is determined predominantly by molecular structure.

A. Rassabina, V. Safronov, Maxim V. Fedorov · 0 citations
Open access Sep 2026

Interpretable Machine Learning Frameworks for QSAR Modeling: A Comparative Study of Classical Algorithms and Transformer-CNN Architectures

An interpretable Quantitative Structure–Activity Relationship (QSAR) framework that integrates classical machine learning approaches with a hybrid deep learning architecture to model enzyme inhibition using a curated and diverse data set from ChEMBL is developed.

Fatima Ezzahra Belkhanchi, M. Mbarki, S. Chtita et al. · 0 citations
Oct 2026

SilicoSAR: An integrated machine-learning QSAR module within the SilicoXplore platform for QSAR-based virtual screening, demonstrated by the discovery of natural-product EGFR inhibitors.

Quantitative structure-activity relationship (QSAR) modelling is central to computer-aided drug discovery, yet many existing QSAR tools operate as isolated descriptor-generation or single-algorithm utilities that require programming expertise and manual transfer of outputs into separate docking and simulation programs....

M. Islam, S. Bhowmick, Sonali Chikhale · 0 citations
#explainable ai Open access Sep 2026

Fragment-Based Explainable AI for Pathogen-Selective Antimicrobial Drug Design

The developed XAI framework, together with the pathogen-specific predictive models, the pathogen-selective fragment lists, and their mapping onto the chemical structures, is able to provide a highly valuable and practical guide to the rational design of new antimicrobial agents.

Abdulmujeeb T. Onawole, M. Blaskovich, Johannes Zuegg · 0 citations
#small language model Open access Sep 2026

Language of Toxicity: An eXplainable Artificial Intelligence Approach

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,...

N. Amoroso, E. Pantaleo, Fulvio Ciriaco et al. · 0 citations

Related blog posts

Google DeepMind Blog Nov 18, 2026

Genie 4 is now available

We are rolling out a new Project Genie experience to eligible Google AI Ultra subscribers, opening access to our new Genie 4 world model.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.