Pre-training with Fractional Denoising to Enhance Molecular Property Prediction

TL;DR

This paper introduces fractional denoising (Frad), a novel molecular pre-training framework that customizes noise to incorporate chemical priors, significantly improving molecular property prediction by better modeling physical principles.

Contribution

The paper proposes a new fractional denoising approach that decouples noise design from force learning constraints, enabling incorporation of chemical priors for enhanced molecular distribution modeling.

Findings

Outperforms existing pre-training methods across multiple molecular tasks.

Achieves state-of-the-art results in force prediction, quantum properties, and binding affinity.

Refined noise design improves force accuracy and sampling coverage.

Abstract

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.