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

By Dejun Jiang, Zhenxing Wu, Chang-Yu Hsieh, Guangyong Chen, Ben Liao, Zhe Wang, Chao Shen, Dongsheng Cao, Jian Wu, Tingjun Hou. Journal of Cheminformatics, 2020

616 citations recorded by the source corpus.

Graph neural networks (GNN) has been considered as an attractive modelling method for molecular property prediction, and numerous studies have shown that GNN could yield more promising results than traditional descriptor-based methods. In this study, based on 11 public datasets covering various property endpoints, the predictive capacity and computational efficiency of the prediction models developed by eight machine learning (ML) algorithms, including four descriptor-based models (SVM, XGBoost, RF and DNN) and four graph-based models (GCN, GAT, MPNN and Attentive FP), were extensively tested and compared. The results demonstrate that on average the descriptor-based models outperform the graph-based models in terms of prediction accuracy and computational efficiency. SVM generally achieves the best predictions for the regression tasks. Both RF and XGBoost can achieve reliable predictions for the classification tasks, and some of the graph-based models, such as Attentive FP and GCN, can yield outstanding performance for a fraction of larger or multi-task datasets. In terms of computational cost, XGBoost and RF are the two most efficient algorithms and only need a few seconds to train a model even for a large dataset. The model interpretations by the SHAP method can effectively explore the established domain knowledge for the descriptor-based models. Finally, we explored use of these models for virtual screening (VS) towards HIV and demonstrated that different ML algorithms offer diverse VS profiles. All in all, we believe that the off-the-shelf descriptor-based models still can be directly employed to accurately predict various chemical endpoints with excellent computability and interpretability.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/08257eb1faa19c29ddcc31d7d749c9bd262213c5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Materials Chemistry

In Molecular Biology

Citations in the corpus

Everything is Connected: Graph Neural Networks (2023, background): “(cid:76) A canonical example is classifying molecules for their quantum-chemical properties (Gilmer et al., 2017), estimating pharmacological properties like toxicity or solubility (Duvenaud et al., 2015; Xiong et al., 2019; Jiang et al., 2021) or virtual drug screening (Stokes et al., 2020).”
Hopfield Networks is All You Need (2020, methodology): “Predictive performance (ROCAUC) on test set as reported by Jiang et al. (2020) for 50 random splits”
MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design (2022, methodology): “The representation methods on ﬁngerprints include random forest, XGBoost, multi-layer perceptron, etc. Recent work [21], [60], [61], [62] have empirically veriﬁed the effectiveness of descriptors.”
BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation (2021, background): “Graph neural networks (GNNs) have received extensive attention from researchers due to their excellent performance on various graph learning tasks such as social analysis [24, 17, 29], drug discovery [12, 25], traffic forecasting [18, 3, 6], recommendation system [38, 32] and computer vision [39, 4]”
Pre-training Molecular Graph Representation with 3D Geometry (2021, neutral): “[44] Dejun Jiang, Zhenxing Wu, Chang-Yu Hsieh, Guangyong Chen, Ben Liao, Zhe Wang, Chao Shen, Dongsheng Cao, Jian Wu, and Tingjun Hou.”