Abstract:Abstract Accurate predictions of druggability and bioactivities of compounds are desirable to reduce the high cost and time of drug discovery. After more than five decades of continuing developments, quantitative structure–activity relationship (QSAR) methods have been established as indispensable tools that facilitate fast, reliable and affordable assessments of physicochemical and biological properties of compounds in drug-discovery programs. Currently, there are mainly two types of QSAR methods, descriptor-based methods and graph-based methods. The former is developed based on predefined molecular descriptors, whereas the latter is developed based on simple atomic and bond information. In this study, we presented a simple but highly efficient modeling method by combining molecular graphs and molecular descriptors as the input of a modified graph neural network, called hyperbolic relational graph convolution network plus (HRGCN+). The evaluation results show that HRGCN+ achieves state-of-the-art performance on 11 drug-discovery-related datasets. We also explored the impact of the addition of traditional molecular descriptors on the predictions of graph-based methods, and found that the addition of molecular descriptors can indeed boost the predictive power of graph-based methods. The results also highlight the strong anti-noise capability of our method. In addition, our method provides a way to interpret models at both the atom and descriptor levels, which can help medicinal chemists extract hidden information from complex datasets. We also offer an HRGCN+'s online prediction service at https://quantum.tencent.com/hrgcn/.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to solve several key challenges in Quantitative Structure - Activity Relationship (QSAR) modeling during the drug discovery process. Specifically, the authors propose an improved graph neural network method that combines molecular graphs and molecular descriptors as inputs, named Hyperbolic Relational Graph Convolution Network Plus (HRGCN+). The following are the main problems that this paper attempts to solve: 1. **Improve the prediction accuracy of QSAR models**: - Traditional QSAR methods face two main limitations in the big data era: First, molecular descriptors are predefined and cannot be updated automatically; second, it is difficult to select the descriptors that are most relevant to specific properties. Moreover, as the size of the dataset increases, descriptor - based methods may be affected by over - fitting or feature redundancy. - Although Graph Neural Networks (GNNs) can directly learn representations from molecular graphs, they may perform poorly on small datasets due to insufficient data. 2. **Combine the advantages of graph representation and descriptors**: - HRGCN+ combines molecular graphs and molecular descriptors to take advantage of both. Molecular graphs provide an intuitive interpretation at the atomic and bond levels, while molecular descriptors provide additional chemical information, which helps to improve the generalization ability of the model, especially on small datasets. 3. **Enhance the noise resistance and interpretability of the model**: - In drug discovery datasets, noise is inevitable. HRGCN+ shows stronger noise resistance and can handle noisy data while maintaining high prediction performance. - In addition, HRGCN+ provides a method to interpret the model at the atomic and descriptor levels, helping drug chemists extract valuable information from complex data. 4. **Implement an efficient and easy - to - use QSAR modeling tool**: - The paper not only shows the excellent performance of HRGCN+ on multiple drug discovery - related datasets but also provides an online prediction service, making it convenient for researchers to use this method for QSAR modeling. In summary, by proposing HRGCN+, this paper aims to solve the limitations of existing QSAR methods in the big data era, improve prediction accuracy, noise resistance, and interpretability, and provide an efficient and easy - to - use tool for drug discovery.

Hyperbolic relational graph convolution networks plus: a simple but highly efficient QSAR-modeling method

Quantitative Structure–activity Relationship: Promising Advances in Drug Discovery Platforms

QSAR Study on Estrogenic Activity of Structurally Diverse Compounds Using Generalized Regression Neural Network

An Adaptive Graph Learning Method for Automated Molecular Interactions and Properties Predictions

Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models

Comprehensive ensemble in QSAR prediction for drug discovery

A Deep Learning-Based Chemical System for QSAR Prediction

On the QSAR-based cardiotoxicity modeling: with expressiveness-enhanced graph learning model and dual-threshold scheme

A survey of drug-target interaction and affinity prediction methods via graph neural networks

Accurate predictions of drugs aqueous solubility via deep learning tools

A general hypergraph learning algorithm for drug multi-task predictions in micro-to-macro biomedical networks

Graph signal processing based nonlinear QSAR/QSPR model learning for compounds

Uni-QSAR: an Auto-ML Tool for Molecular Property Prediction

CRNNTL: Convolutional Recurrent Neural Network and Transfer Learning for QSAR Modeling in Organic Drug and Material Discovery

Predicting associations between drugs and G protein-coupled receptors using a multi-graph convolutional network

AI-enhanced chemical paradigm: From molecular graphs to accurate prediction and mechanism

Neural networks in QSAR studies: Estimation and prediction of biological activity for N-(substituted phenoxyethyl)-cyclopropylamines

Drug–target affinity prediction with extended graph learning-convolutional networks

Development and Evaluation of Conformal Prediction Methods for QSAR

YZS-model: A Predictive Model for Organic Drug Solubility Based on Graph Convolutional Networks and Transformer-Attention