Toxic molecular compounds can be identified early, which can lower drug development costs and avoid failures in candidate screening. However, current machine learning methods typically require the creation of handcrafted molecular descriptors and are not effective at modelling complex molecular relationships. In this study, a hybrid quantum-classical molecular toxicity prediction framework is proposed to be used for drug candidate screening, based on a Variational Quantum Eigensolver (VQE). SMILES strings are used to represent molecular structures which are processed in RDKit to produce physicochemical descriptors, and quantum chemical properties from the QM9 dataset are used to enrich the molecular representation. The reduced features are then encoded into a 12-qubit variational quantum circuit for binary toxicity classification after feature standardization and Principal Component Analysis (PCA) was performed. The trained VQE-inspired classifier is deployed via a web application built using Flask with facilities for authenticated predictions of toxicity, visualization of molecular properties, and management of prediction history. The experimental results show that it is feasible to combine molecular feature engineering with hybrid quantum learning to obtain an overall classification accuracy of 84% and a recall of 94% of the toxicclass molecules. The proposed framework gives a computational platform from end to end for quantum assisted molecular toxicity prediction and provides a foundation of future quantum assisted computer-aided drug discovery.
Toxic molecular compounds can be identified early, which can lower drug development costs and avoid failures in candidate screening. However, current machine learning methods typically require the creation of handcrafted molecular descriptors and are not effective at modelling complex molecular relationships. In this study, a hybrid quantum-classical molecular toxicity prediction framework is proposed to be used for drug candidate screening, based on a Variational Quantum Eigensolver (VQE). SMILES strings are used to represent molecular structures which are processed in RDKit to produce physicochemical descriptors, and quantum chemical properties from the QM9 dataset are used to enrich the molecular representation. The reduced features are then encoded into a 12-qubit variational quantum circuit for binary toxicity classification after feature standardization and Principal Component Analysis (PCA) was performed. The trained VQE-inspired classifier is deployed via a web application built using Flask with facilities for authenticated predictions of toxicity, visualization of molecular properties, and management of prediction history. The experimental results show that it is feasible to combine molecular feature engineering with hybrid quantum learning to obtain an overall classification accuracy of 84% and a recall of 94% of the toxicclass molecules. The proposed framework gives a computational platform from end to end for quantum assisted molecular toxicity prediction and provides a foundation of future quantum assisted computer-aided drug discovery.
Classical computational approaches are steadily improving in their ability to predict molecular properties based on their sequences or structure, which is essential in drug discovery, irrespective of its many challenges. The major challenge is to use computational approaches to identify the properties, such as toxicity, solubility, etc. (ADMET), of the drug targets without costing much time. Machine learning and deep learning methods are employed for predicting ADMET properties. The proposed approach combines quantum computing with machine learning for the binary task of toxicity prediction based on the SMILES data. This Quantum Machine Learning (QML) approach involves a novel five-step procedure that converts SMILES to their respective molecular fingerprints, which are in turn dimensionally reduced, consequently forming four qubits. These qubits are inputs to the QML models, such as the Quantum SVC (QSVC), Variational Quantum Circuits (VQC), and Quantum AutoEncoders (QAE). The models are trained and optimized by using different optimizers and then evaluated based on the accuracy metric. The QAE model outperformed the remaining model by achieving an accuracy of 99%.
The pharmaceutical industry faces mounting challenges in drug discovery, where exploring a chemical space exceeding 10 60 possible compounds remains computationally prohibitive. Traditional in-silico screening approaches struggle to capture the quantum mechanical interactions that determine molecular behavior. To overcome these limitations, this study introduces a hybrid quantum–classical machine learning framework that uniquely integrates molecular graph theory with quantum-enhanced learning for accelerated drug discovery. Using the QM9 benchmark dataset containing 133,885 molecules of low-to-medium structural complexity, we extract Laplacian eigenvalue spectra as topological descriptors that preserve both connectivity and electronic information. These features are jointly processed by Random Forest (RF), Support Vector Machine (SVM), and a Variational Quantum Classifier (VQC) in a weighted ensemble architecture designed for balanced generalization. The Quantum Graph Neural Network (QGNN) component achieves 97.5% accuracy, while the hybrid ensemble attains 100.0% accuracy, confirming strong predictive reliability. Although quantum training requires roughly 3.2× longer computation, it captures high-dimensional correlations that classical methods overlook. Overall, the proposed framework bridges the gap between theoretical quantum computing and practical pharmaceutical modeling, providing a scalable, physics-inspired pathway toward intelligent molecular screening and next-generation drug discovery systems.
Abstract: This study explores the implementation of hybrid quantum-classical algorithms in the realm of drug discovery and molecular simulation, with an emphasis on optimizing molecular interaction prediction. The integration of quantum computing paradigms, particularly the Variational Quantum Eigensolver (VQE), with classical machine learning techniques is evaluated for its efficacy in simulating molecular systems. Key results demonstrate superior accuracy in predicting binding affinities and reaction energetics using the hybrid models compared to traditional methods. The findings offer significant implications for accelerating the drug development pipeline, particularly in early-stage candidate screening. Keywords: Hybrid computing, Quantum simulation, Molecular interaction, Drug discovery, Variational Quantum Eigensolver, Quantum-classical integration
In the drug discovery paradigm, the evaluation of absorption, distribution, metabolism, and excretion (ADME) and toxicity properties of new chemical entities is one of the most critical issues, which is a time-consuming process, immensely expensive, and poses formidable challenges in pharmaceutical R&D. In recent years, emerging technologies like artificial intelligence (AI), big data, and cloud technologies have garnered great attention to predict the ADME and toxicity of molecules. Currently, the blend of quantum computation and machine learning has attracted considerable attention in almost every field ranging from chemistry to biomedicine and several engineering disciplines as well. Quantum computers have the potential to bring advances in high-throughput experimental techniques and in screening billions of molecules by reducing development costs and time associated with the drug discovery process. Motivated by the efficiency of quantum kernel methods, we proposed a quantum machine learning (QML) framework consisting of a classical support vector classifier algorithm with a kernel-based quantum classifier. To demonstrate the feasibility of the proposed QML framework, the simplified molecular input line entry system (SMILES) notation-based string kernel, combined with a quantum support vector classifier, is used for the evaluation of chemical/drug ADME-Tox properties. The proposed quantum machine learning framework is validated and assessed via large-scale simulations. Based on our results from numerical simulations, the quantum model achieved the best performance as compared to classical counterparts in terms of the area under the curve of the receiver operating characteristic curve (AUC ROC; 0.80-0.95) for predicting outcomes on ADME-Tox data sets for small molecules, with a different number of features. The deployment of the proposed framework in the pharmaceutical industry would be extremely valuable in making the best decisions possible.
Abstract Quantum computing has emerged as a powerful tool for modeling molecular systems with high accuracy, accelerating the search for novel therapeutic targets. In this study, we present a two-stage hybrid workflow that combines a Quantum Graph Neural Network with a Variational Quantum Eigensolver to identify potent serine neutralizers in the QM9 dataset. In the first stage, an advanced quantum graph neural network architecture incorporating attention layers, self-distillation, and an adaptive learning-rate schedule is trained to predict ionization potentials and binding free energies. Across five independent random-seed trials, our adaptive-thresholded QGNN–VQE pipeline achieves an average R2 R 2 of 0.990±0.008 0.990 ± 0.008 and a mean absolute error of 0.034±0.001 0.034 ± 0.001 eV ( ≈0.79±0.03 ≈ 0.79 ± 0.03 kcal/mol) on the QM9 validation set, demonstrating robust, chemical-accuracy-level predictions. Building on these high-precision predictions, the second stage employs a QAOA-inspired hybrid ranking scheme that merges quantum graph neural network outputs, feature-space similarity (via and cosine similarity), and variational quantum eigensolver-derived energy stability. This α α -weighted ( α=0.95 α = 0.95 ) scoring framework is applied to over 133,000 molecules, efficiently parsed in parallel. The final ranked list is validated through an improved bijection test, featuring an adaptive similarity threshold to ensure both structural and functional robustness. The top-scoring candidate identified by our pipeline is 5,6,7-tetrahydro-4 H-pyrazolo[4,3-c]pyridin-4-one, highlighting the efficacy of quantum-enhanced modeling in pinpointing complex pharmacological targets. These findings underscore the transformative potential of hybrid quantum-classical workflows in drug discovery, offering a flexible blueprint for advanced screening of other critical biomolecular interactions in the rapidly expanding field of quantum biology. GraphicAbstract VQE for Serine Neutralization
Quantum Computing (QC) offers outstanding potential for molecular characterization and drug discovery, particularly in solving complex properties like the Ground State Energy (GSE) of biomolecules. However, QC faces challenges due to computational noise, scalability, and system complexity. This work presents a hybrid framework combining Machine Learning (ML) techniques with quantum algorithms−Variational Quantum Eigensolver (VQE), Hartree-Fock (HF), and Quantum Phase Estimation (QPE)−to improve GSE predictions for large molecules. Three datasets (chemical descriptors, Coulomb matrices, and a hybrid combination) were prepared using molecular features from PubChem. These datasets trained XGBoost (XGB), Random Forest (RF), and LightGBM (LGBM) models. XGB achieved the lowest Relative Error (RE) of 4.41±11.18% on chemical descriptors, outperforming RF (5.56±11.66%) and LGBM (5.32±12.87%). HF delivered exceptional precision for small molecules (0.44±0.66%RE), while a near-linear correlation between GSE and molecular electron count provided predictive shortcuts. This study demonstrates that integrating QC and ML enhances scalability for molecular energy predictions and lays the foundation for scaling QC molecular simulations to larger systems.