graph neural networks

OpenFOAM data of the lid-driven cavity, natural-convection cavity, heated cylinder wake and data center benchmarks

OpenFOAM data of the benchmarks of the manuscript "Label-Free Finite-Volume-Residual Training of Attention Graph Neural Networks for Coupled Thermo-Fluid Fields".

Validating and Correcting Graph Neural Network Attention for Drug-Induced Liver Injury Prediction

Abstract Attention-based graph neural networks (GNNs) are increasingly used for drug-induced liver injury (DILI) and other toxicity prediction tasks on the claimed strength of built-in interpretability, but this claim is almost always supported by a handful of hand-selected examples rather than being tested systematica...

ReactionGraphNeuralNetwork: graph neural network surrogate models for reaction properties

Code for graph neural network-based surrogate models used to predict reaction-related properties for reinforcement-learning-accelerated atomistic simulations.

Using Graph Neural Networks (GNN) and Gated Recurrent Unit (GRU) based on Regression Modelling for Climatic Forecasting

AChE Molecular Machine Learning: Structure-Aware Prediction of Acetylcholinesterase Inhibitory Potency

A structure-aware molecular machine-learning pipeline for continuous prediction of acetylcholinesterase inhibitory potency from molecular structure. The project includes ChEMBL activity retrieval and curation, IC50 normalization and pIC50 transformation, Morgan fingerprints, RDKit molecular descriptors, scaffold-aware...

AChE Molecular Machine Learning: Structure-Aware Prediction of Acetylcholinesterase Inhibitory Potency

A structure-aware molecular machine-learning pipeline for continuous prediction of acetylcholinesterase inhibitory potency from molecular structure. The project includes ChEMBL activity retrieval and curation, IC50 normalization and pIC50 transformation, Morgan fingerprints, RDKit molecular descriptors, scaffold-aware...

Gated-Slot Linear Attention Graph Neural Networks for Session-based Recommendation

Expressivity of quantum graph neural networks with variable-dimension edge qudits

A Graph Neural Network for Global Daily Fire Radiative Power Prediction at Medium-Range Lead Times

The model is a lightweight graph neural network. It treats each spatial patch as a grid graph, applies residual graph convolutions to each day of input history, and then aggregates the history with temporal convolutions. A two-part (hurdle) output head predicts both the probability of occurrence and the conditional int...

AI applications in drug discovery and personalized medicine

The era of artificial intelligence (AI) in drug discovery and personalized medicine is bringing a new twist to the healthcare field, enhancing the ability to identify the target, develop drugs, and design personal treatment regimens when taking into account the profile of a particular patient. They allow for incorporat...

Graph-Based Fault Localization in Microservices Using Static Code Metrics

AChE Molecular Machine Learning: Structure-Aware Prediction of Acetylcholinesterase Inhibitory Potency

A structure-aware molecular machine-learning pipeline for continuous prediction of acetylcholinesterase inhibitory potency from molecular structure. The project includes ChEMBL activity retrieval and curation, IC50 normalization and pIC50 transformation, Morgan fingerprints, RDKit molecular descriptors, scaffold-aware...