A5043971485- Machine Learning in Materials Science
- Computational Drug Discovery Methods
- Luminescence Properties of Advanced Materials
- Metabolomics and Mass Spectrometry Studies
- Radiation Detection and Scintillator Technologies
- Protein Structure and Dynamics
- Inorganic Fluorides and Related Compounds
Public Health Ontario
2025
University of Toronto
2024-2025
National Research Tomsk State University
2016
Tomsk Polytechnic University
2016
Abstract Motivation Drug–target interaction (DTI) prediction aims to identify interactions between drugs and protein targets. Deep learning can automatically learn discriminative features from drug target representations for DTI prediction, but challenges remain, making it an open question. Existing approaches encode targets into using deep models, they often lack explanations underlying interactions. Moreover, limited labeled DTIs in the chemical space hinder model generalization. Results...
Predicting molecular properties is essential for drug discovery, and computational methods can greatly enhance this process. Molecular graphs have become a focus representation learning, with Graph Neural Networks (GNNs) widely used. However, GNNs often struggle capturing long-range dependencies. To address this, we propose MolGraph-xLSTM, novel graph-based xLSTM model that enhances feature extraction effectively models molecule interactions. Our approach processes at two scales: atom-level...
Understanding compound-protein interactions is crucial for early drug discovery, offering insights into molecular mechanisms and potential therapeutic effects of compounds. Here, we introduce GraphBAN, a graph-based framework that inductively predicts these using compound protein feature information. GraphBAN effectively handles inductive link predictions unseen nodes, providing robust solution predicting between entirely compounds proteins. This capability enables to transcend the...
The spectra and kinetics of pulsed cathodoluminescence buildup decay in LiF-TiO 2 LiF-WO 3. crystals have been studied the temperature range 20–300 K. It is found that all LiF doped with metal oxides (Fe, Ti, W) a similar structure luminescence centers – oxygen excited by impurities Fe O 3 , WO TiO complexes. In types crystals, processes energy transfer to are (buildup stage).