# FragNet: A Graph Neural Network for Molecular Property Prediction with Four Levels of Interpretability

arXiv: https://arxiv.org/abs/2410.12156 · HTML: https://mlsanity.com/paper/2410.12156/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 61 references that are in the map.
- First posted on arXiv 16 Oct 2024.

## Builds on (most cited first, 19 of 19)

- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Neural Message Passing for Quantum Chemistry](https://mlsanity.com/paper/1704.01212/) (2017)
- [Towards A Rigorous Science of Interpretable Machine Learning](https://mlsanity.com/paper/1702.08608/) (2017)
- [Strategies for Pre-training Graph Neural Networks](https://mlsanity.com/paper/1905.12265/) (2019)
- [GNNExplainer: Generating Explanations for Graph Neural Networks](https://mlsanity.com/paper/1903.03894/) (2019)
- [The Mythos of Model Interpretability](https://mlsanity.com/paper/1606.03490/) (2016)
- [Analyzing Learned Molecular Representations for Property Prediction](https://mlsanity.com/paper/1904.01561/) (2019)
- [Graph Contrastive Learning Automated](https://mlsanity.com/paper/2106.07594/) (2021)
- [Pre-training Molecular Graph Representation with 3D Geometry](https://mlsanity.com/paper/2110.07728/) (2021)
- [ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction](https://mlsanity.com/paper/2106.06130/) (2021)
- [Motif-based Graph Self-Supervised Learning for Molecular Property Prediction](https://mlsanity.com/paper/2110.00987/) (2021)
- [Rethinking Tokenizer and Decoder in Masked Graph Modeling for Molecules](https://mlsanity.com/paper/2310.14753/) (2023)
- [MEG: Generating Molecular Counterfactual Explanations for Deep Graph Networks](https://mlsanity.com/paper/2104.08060/) (2021)
- [Exploring QSAR Models for Activity-Cliff Prediction](https://mlsanity.com/paper/2301.13644/) (2023)
- [ProGReST: Prototypical Graph Regression Soft Trees for Molecular Property Prediction](https://mlsanity.com/paper/2210.03745/) (2022)
- [Explainability Techniques for Chemical Language Models](https://mlsanity.com/paper/2305.16192/) (2023)
- [Explainable Molecular Property Prediction: Aligning Chemical Concepts with Predictions via Language Models](https://mlsanity.com/paper/2405.16041/) (2024)

## Built on by (most cited first, 3 of 3)

- [MolEvolve: LLM-Guided Evolutionary Search for Interpretable Molecular Optimization](https://mlsanity.com/paper/2603.24382/) (2026)
- [BiScale-GTR: Fragment-Aware Graph Transformers for Multi-Scale Molecular Representation Learning](https://mlsanity.com/paper/2604.06336/) (2026)
- [Multi-level Self-supervised Pretraining on Compositional Hierarchical Graph for Molecular Property Prediction](https://mlsanity.com/paper/2605.16088/) (2026)

## Abstract (arXiv)

Molecular property prediction is essential in a variety of contemporary scientific fields, such as drug development and designing energy storage materials. Although there are many machine learning models available for this purpose, those that achieve high accuracy while also offering interpretability of predictions are uncommon. We present a graph neural network that not only matches the prediction accuracies of leading models but also provides insights on four levels of molecular substructures. This model helps identify which atoms, bonds, molecular fragments, and connections between fragments are significant for predicting a specific molecular property. Understanding the importance of connections between fragments is particularly valuable for molecules with substructures that do not connect through standard bonds. The model additionally can quantify the impact of specific fragments on the prediction, allowing the identification of fragments that may improve or degrade a property value. These interpretable features are essential for deriving scientific insights from the model's learned relationships between molecular structures and properties.