Geometry-Informed Parameter-Efficient Fine-Tuning of Pre-trained Molecular GNNs for Blood-Brain Barrier Permeability Prediction

2026-08-04

AI Summary

This work addresses key challenges in blood–brain barrier permeability (BBBP) prediction—namely data scarcity, class imbalance, and high sensitivity to molecular three-dimensional structure—by proposing the BBBP-GeoPEFT framework. It introduces 3D geometric information into parameter-efficient fine-tuning for the first time, modeling spatial interactions between atoms and second-order edges via multi-cutoff distance graphs and line graphs. A lightweight geometric encoder is designed, incorporating node-level cutoff attention and gated residual connections to enable efficient adaptation. Remarkably, by updating only 10.1% of the model parameters, BBBP-GeoPEFT matches or exceeds the performance of full fine-tuning under both random and scaffold splits, significantly outperforming existing parameter-efficient fine-tuning approaches.

Abstract

Blood-brain barrier permeability (BBBP) prediction is a critical screening task in central nervous system drug discovery, where candidate molecules must be assessed for whether they can cross, or should be prevented from crossing, the blood-brain barrier. However, this task remains challenging because of limited, class-imbalanced datasets and sensitivity to molecular structure. Recent advances in deep learning have established graph neural networks (GNNs) as a powerful approach for molecular representation learning, while pre-trained molecular GNNs provide transferable knowledge for downstream tasks. However, full fine-tuning is often parameter-inefficient and prone to overfitting, whereas existing parameter-efficient fine-tuning (PEFT) methods mainly adapt node features or the two-dimensional covalent graph, limiting their ability to capture three-dimensional geometry and second-order interactions. To address these limitations, we propose BBBP-GeoPEFT, a geometry-informed PEFT framework for pre-trained molecular GNNs. BBBP-GeoPEFT constructs distance-based graphs at multiple cutoffs and their corresponding line graphs from molecular conformers to capture spatial atom and second-order edge interactions. Lightweight auxiliary geometric graph encoders generate cutoff-specific representations, which are incorporated into each pre-trained layer through node-wise cutoff attention and gated residual connections. This design preserves pre-trained knowledge while incorporating permeability-relevant geometric information with a small trainable-parameter budget. Experiments on a curated BBBP dataset show that BBBP-GeoPEFT achieves competitive performance compared with full fine-tuning and representative PEFT baselines. Under both random and scaffold splitting, BBBP-GeoPEFT achieves competitive or improved ROC-AUC and accuracy in most experiments while updating only 10.1% of the model parameters.

Problem

• blood-brain barrier permeability
• molecular structure
• class imbalance
• limited data
• 3D geometry

Innovation

• geometry-informed fine-tuning
• parameter-efficient fine-tuning
• molecular GNNs
• 3D molecular representation
• blood-brain barrier permeability