On Improving Graph Neural Networks for QSAR by Pre-training on Extended-Connectivity Fingerprints
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910209418461184 |
|---|---|
| author | Money-Kyrle, Sam Dablander, Markus Hanser, Thierry Werner, Stephane Deane, Charlotte M. Morris, Garrett M. |
| author_facet | Money-Kyrle, Sam Dablander, Markus Hanser, Thierry Werner, Stephane Deane, Charlotte M. Morris, Garrett M. |
| contents | Molecular Graph Neural Networks (GNNs) are increasingly common in drug discovery, particularly for Quantitative Structure-Activity Relationship (QSAR) studies; yet, their superiority compared to classical molecular featurisation approaches is disputed. We report a general strategy for improving GNNs for QSAR by pre-training to predict Extended-Connectivity Fingerprints (ECFP). We validate our approach with statistical tests and challenging out-of-distribution (OOD) splits. Across five out of six Biogen benchmarks, we observed a statistically significant improvement in standard performance metrics over all evaluated baselines when using ECFP pre-trained GNNs. However, for more heterogeneous datasets and more complex endpoints, such as binding affinity prediction, pre-trained GNNs underperformed in OOD settings. Importantly, we investigated the impact of substructure-level data leakage during pre-training on downstream performance. While we identified scenarios where pre-training on ECFPs was less effective, our findings show that ECFP-based pre-training can enhance downstream OOD performance on a diverse set of practically relevant QSAR tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_10722 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | On Improving Graph Neural Networks for QSAR by Pre-training on Extended-Connectivity Fingerprints Money-Kyrle, Sam Dablander, Markus Hanser, Thierry Werner, Stephane Deane, Charlotte M. Morris, Garrett M. Machine Learning Molecular Graph Neural Networks (GNNs) are increasingly common in drug discovery, particularly for Quantitative Structure-Activity Relationship (QSAR) studies; yet, their superiority compared to classical molecular featurisation approaches is disputed. We report a general strategy for improving GNNs for QSAR by pre-training to predict Extended-Connectivity Fingerprints (ECFP). We validate our approach with statistical tests and challenging out-of-distribution (OOD) splits. Across five out of six Biogen benchmarks, we observed a statistically significant improvement in standard performance metrics over all evaluated baselines when using ECFP pre-trained GNNs. However, for more heterogeneous datasets and more complex endpoints, such as binding affinity prediction, pre-trained GNNs underperformed in OOD settings. Importantly, we investigated the impact of substructure-level data leakage during pre-training on downstream performance. While we identified scenarios where pre-training on ECFPs was less effective, our findings show that ECFP-based pre-training can enhance downstream OOD performance on a diverse set of practically relevant QSAR tasks. |
| title | On Improving Graph Neural Networks for QSAR by Pre-training on Extended-Connectivity Fingerprints |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.10722 |