Abstract:Determining various properties of molecules is a critical step in drug discovery. Recently, with the improvement of large heterogeneous datasets and the development of deep learning approaches, more and more scientists have turned their attention to neural network-based virtual preliminary screening to reduce the time and monetary cost of drug discovery. However, the poor interpretability of deep learning masks causality, so models' conclusions are often beyond the comprehension of human users, which reduces the credibility of the model and makes it difficult for chemists to further narrow the huge chemical space based on models' results. Thus, this study develops a novel framework consisting of Graph Neural Networks for feature extraction, Curriculum-Based Learning Strategies for optimization, and a Learning Binary Neural Tree (LBNT) for prediction, to improve the performance of neural networks and reveal their decision-making process to chemists. The framework encodes molecular graph data with graph neural networks (GNNs), then retrains the encoder with curriculum-based learning strategies to reduce uncertainty and improve accuracy, and finally uses LBNT as the predictor, which joint retrains with the encoder after independently training, for prediction and visualization. The framework is validated on the public datasets and compared to single GNNs with normal training strategies as well as GNN encoders with common machine learning predictors instead of the LBNT predictor. The result reveals that the proposed framework enhances the point prediction accuracy of the completely trained GNN and reduces its uncertainty through curriculum-based learning, and further improves the accuracy by combining LBNT. Besides, compared with common machine learning tools, the LBNT predictor generally has the best performance because of joint retraining with the GNN encoder. The decision-making process of LBNT is also better and easier to explain than that of other models.

Graph neural networks with molecular segmentation for property prediction and structure–property relationship discovery

Graph neural networks for the prediction of molecular structure-property relationships

HiGNN: Hierarchical Informative Graph Neural Networks for Molecular Property Prediction Equipped with Feature-Wise Attention

Multi-View Graph Neural Networks for Molecular Property Prediction

Cross-dependent graph neural networks for molecular property prediction

Chemistry-intuitive explanation of graph neural networks for molecular property prediction with substructure masking

A drug molecular classification model based on graph structure generation

Describe Molecules by a Heterogeneous Graph Neural Network with Transformer-like Attention for Supervised Property Predictions

Introducing Block Design in Graph Neural Networks for Molecular Properties Prediction

Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Properties

Graph Neural Network Architecture Search for Molecular Property Prediction

Graph neural network for 3-dimensional structures including dihedral angles for molecular property prediction

MD-GNN: A mechanism-data-driven graph neural network for molecular properties prediction and new material discovery

Graph Neural Tree: A novel and interpretable deep learning-based framework for accurate molecular property predictions

Molecular Property Prediction Based on Graph Structure Learning

Graph Neural Networks for Molecules

Enhancing property and activity prediction and interpretation using multiple molecular graph representations with MMGX

Enhancing Model Learning and Interpretation Using Multiple Molecular Graph Representations for Compound Property and Activity Prediction

Chain-aware graph neural networks for molecular property prediction

Graph Neural Network-Based Molecular Property Prediction with Patch Aggregation

Quantitative evaluation of explainable graph neural networks for molecular property prediction