Pre-training with Fractional Denoising to Enhance Molecular Property Prediction
This work addresses a key bottleneck in molecular property prediction for drug discovery and material design by improving pre-training with chemical priors, though it is incremental as it builds on existing denoising methods.
The paper tackled the problem of bias in molecular pre-training due to limited noise distributions by introducing a fractional denoising framework that decouples noise design from force learning constraints, resulting in state-of-the-art performance across force prediction, quantum chemical properties, and binding affinity tasks.
Deep learning methods have been considered promising for accelerating molecular screening in drug discovery and material design. Due to the limited availability of labelled data, various self-supervised molecular pre-training methods have been presented. While many existing methods utilize common pre-training tasks in computer vision (CV) and natural language processing (NLP), they often overlook the fundamental physical principles governing molecules. In contrast, applying denoising in pre-training can be interpreted as an equivalent force learning, but the limited noise distribution introduces bias into the molecular distribution. To address this issue, we introduce a molecular pre-training framework called fractional denoising (Frad), which decouples noise design from the constraints imposed by force learning equivalence. In this way, the noise becomes customizable, allowing for incorporating chemical priors to significantly improve molecular distribution modeling. Experiments demonstrate that our framework consistently outperforms existing methods, establishing state-of-the-art results across force prediction, quantum chemical properties, and binding affinity tasks. The refined noise design enhances force accuracy and sampling coverage, which contribute to the creation of physically consistent molecular representations, ultimately leading to superior predictive performance.