MolCPT: Molecule Continuous Prompt Tuning to Generalize Molecular Representation Learning

TL;DR

MolCPT introduces a novel continuous prompt tuning method leveraging motif subgraphs to enhance molecular property prediction, significantly improving the generalization of pre-trained GNNs with minimal fine-tuning.

Contribution

The paper proposes MolCPT, a new paradigm that incorporates motif-based continuous prompts into pre-trained GNNs to better capture molecular structures for property prediction.

Findings

MolCPT outperforms existing methods on benchmark datasets.

It effectively leverages motif subgraphs for improved generalization.

Minimal fine-tuning is required for strong performance.

Abstract

Molecular representation learning is crucial for the problem of molecular property prediction, where graph neural networks (GNNs) serve as an effective solution due to their structure modeling capabilities. Since labeled data is often scarce and expensive to obtain, it is a great challenge for GNNs to generalize in the extensive molecular space. Recently, the training paradigm of "pre-train, fine-tune" has been leveraged to improve the generalization capabilities of GNNs. It uses self-supervised information to pre-train the GNN, and then performs fine-tuning to optimize the downstream task with just a few labels. However, pre-training does not always yield statistically significant improvement, especially for self-supervised learning with random structural masking. In fact, the molecular structure is characterized by motif subgraphs, which are frequently occurring and influence molecular properties. To leverage the task-related motifs, we propose a novel paradigm of "pre-train, prompt, fine-tune" for molecular representation learning, named molecule continuous prompt tuning (MolCPT). MolCPT defines a motif prompting function that uses the pre-trained model to project the standalone input into an expressive prompt. The prompt effectively augments the molecular graph with meaningful motifs in the continuous representation space; this provides more structural patterns to aid the downstream classifier in identifying molecular properties. Extensive experiments on several benchmark datasets show that MolCPT efficiently generalizes pre-trained GNNs for molecular property prediction, with or without a few fine-tuning steps.