A spatial interface between biological graphs and vision models.

Complex biological networks are fundamental to biomedical science, capturing interactions among molecules, cells, genes, and tissues. Deciphering these networks is critical for understanding health and disease, yet their scale and complexity represent a daunting challenge for current computational methods. Traditional biological network analysis methods, including deep learning approaches, while powerful, face inherent challenges such as limited scalability, oversmoothing long-range dependencies, difficulty in multimodal integration, expressivity bounds, and poor interpretability. Here, we present Graph2Image, a framework that transforms large biological networks into sets of two-dimensional images by spatially arranging representative network nodes on a 2D grid. This transformation decouples the nodes as images, enabling the use of convolutional neural networks with global receptive fields and multi-scale pyramids, thus overcoming limitations of existing biological network analysis methods in scalability, memory efficiency, and long-range context capture. Graph2Image also facilitates seamless integration with other omics modalities and enhances interpretability through direct visualization of node-associated images. When applied to several large-scale biological network datasets, Graph2Image improved classification accuracy by up to 26.2% over the strongest graph neural network baseline. Beyond predictive performance, Graph2Image generated interpretable attributions that captured biologically and therapeutically relevant cancer dependencies, distinguishing cancer-differentiating genes from housekeeping genes. Applied to breast cancer networks, Graph2Image identified a co-expression module associated with basal-like breast cancer and linked it to genes currently being pursued as therapeutic targets. The proposed approach scales linearly with graph size, whereas conventional graph neural networks require iterative message passing over the graph, resulting in substantially higher computational costs for large networks. Graph2Image thus provides a scalable and interpretable approach to biological network analysis that enables the identification of disease-relevant targets and the mechanistic study of complex biological systems.