mlsanityData to 7 Oct 2026

Quantitative Evaluation of Explainable Graph Neural Networks for Molecular Property Prediction

arXiv 2107.04119 ↗First posted 1 Jul 2021q-bio.QM cs.LGcited by 15

Not on the map. Why?

Abstract (arXiv)

Advances in machine learning have led to graph neural network-based methods for drug discovery, yielding promising results in molecular design, chemical synthesis planning, and molecular property prediction. However, current graph neural networks (GNNs) remain of limited acceptance in drug discovery is limited due to their lack of interpretability. Although this major weakness has been mitigated by the development of explainable artificial intelligence (XAI) techniques, the "ground truth" assignment in most explainable tasks ultimately rests with subjective judgments by humans so that the quality of model interpretation is hard to evaluate in quantity. In this work, we first build three levels of benchmark datasets to quantitatively assess the interpretability of the state-of-the-art GNN models. Then we implemented recent XAI methods in combination with different GNN algorithms to highlight the benefits, limitations, and future opportunities for drug discovery. As a result, GradInput and IG generally provide the best model interpretability for GNNs, especially when combined with GraphNet and CMPNN. The integrated and developed XAI package is fully open-sourced and can be used by practitioners to train new models on other drug discovery tasks.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Builds on 32 of its 68 references that are in the map. First posted on arXiv 1 Jul 2021.

Builds on

32 of 68 references.

References · 32

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2017Graph Attention Networkscited by 8,854
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  5. 2017Inductive Representation Learning on Large Graphscited by 6,284
  6. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  7. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  8. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  9. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  10. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  11. 2018Graph Neural Networks: A Review of Methods and Applicationscited by 1,571
  12. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
20 more
  1. 2018Relational inductive biases, deep learning, and graph networkscited by 1,341
  2. 2018Graph Convolutional Neural Networks for Web-Scale Recommender Systemscited by 1,133
  3. 2017SmoothGrad: removing noise by adding noisecited by 1,097
  4. 2018A Survey Of Methods For Explaining Black Box Modelscited by 970
  5. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 735
  6. 2019Graph Neural Networks for Social Recommendationcited by 711
  7. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 647
  8. 2020Parameterized Explainer for Graph Neural Networkcited by 423
  9. 2020Explainability in Graph Neural Networks: A Taxonomic Surveycited by 383
  10. 2019Analyzing Learned Molecular Representations for Property Predictioncited by 373
  11. 2019Unmasking Clever Hans Predictors and Assessing What Machines Really Learncited by 359
  12. 2018Graph networks as learnable physics engines for inference and controlcited by 242
  13. 2018Constrained Graph Variational Autoencoders for Molecule Designcited by 182
  14. 2020Graph Information Bottleneck for Subgraph Recognitioncited by 119
  15. 2020Drug discovery with explainable artificial intelligencecited by 118
  16. 2020RetroXpert: Decompose Retrosynthesis Prediction like a Chemistcited by 66
  17. 2014Understanding Random Forests: From Theory to Practicecited by 57
  18. 2020Exploiting Neuron and Synapse Filter Dynamics in Spatial Temporal Learning of Deep Spiking Neural Networkcited by 36
  19. 2021MEG: Generating Molecular Counterfactual Explanations for Deep Graph Networkscited by 32
  20. 2018Using Attribution to Decode Dataset Bias in Neural Network Models for Chemistrycited by 25

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 3; 2024: 1; 2025: 5; 2026: 3.

15 papers.

Sort:

  1. 2022A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protectioncited by 20
  2. 2022Towards Faithful and Consistent Explanations for Graph Neural Networkscited by 13
  3. 2025Graph Neural Networks in Modern AI-aided Drug Discoverycited by 12
  4. 2022Towards Explainable Motion Prediction using Heterogeneous Graph Representationscited by 10
  5. 2023Faithful and Consistent Graph Neural Network Explanations with Rationale Alignmentcited by 6
  6. 2023Knowledge-augmented Graph Machine Learning for Drug Discovery: A Surveycited by 6
  7. 2023Explainable Artificial Intelligence for Drug Discovery and Development -- A Comprehensive Surveycited by 4
  8. 2025MTPNet: Multi-Grained Target Perception for Unified Activity Cliff Predictioncited by 3
  9. 2026A Semi-supervised Molecular Learning Framework for Activity Cliff Estimationcited by 3
  10. 2025Structure-Aware Compound-Protein Affinity Prediction via Graph Neural Networks with Group Lasso Regularizationcited by 1
  11. 2024Incorporating Retrieval-based Causal Learning with Information Bottlenecks for Interpretable Graph Neural Networkscited by 0
  12. 2025Rotational Sampling: A Plug-and-Play Encoder for Rotation-Invariant 3D Molecular GNNscited by 0
3 more
  1. 2025ARM-Explainer -- Explaining and improving graph neural network predictions for the maximum clique problem using node features and association rule miningcited by 0
  2. 2026Why Self-Inconsistency Arises in GNN Explanations and How to Exploit Itcited by 0
  3. 2026MolLedger: An Additive Graph Neural Network with Chemically Grounded ADME Attributionscited by 0