Gene regulatory networks (GRNs) describe the interactions between transcription factors (TFs) and their target genes, playing a crucial role in understanding gene functions and how cells regulate gene expression under different conditions. Recent advancements in multi-omics technologies have provided new opportunities for more comprehensive GRN inference. Among these data types, gene expression and chromatin accessibility are particularly important, as they are key to distinguishing between direct and indirect regulatory relationships. However, existing methods primarily rely on gene expression data while neglecting biological information such as chromatin accessibility, leading to an increased occurrence of false positives in the inference results. To address the limitations of existing approaches, we propose MultiGNN, a supervised framework based on graph neural networks (GNNs). Unlike conventional GRN inference methods, MultiGNN leverages features extracted from both gene expression and chromatin accessibility data to predict regulatory interactions between genes. Experimental results demonstrate that MultiGNN consistently outperforms other methods across seven datasets. Additionally, ablation studies validate the effectiveness of our multi-omics feature integration strategy, offering a new direction for more accurate GRN inference.

Gene Regulatory Network (GRN) describes the regulatory interactions between transcription factors (TFs) and their target genes, playing a crucial role in understanding developmental biology, disease mechanisms, and drug target discovery. However, due to the complexity of gene regulation, inferring GRNs solely from gene expression data remains highly challenging. Additionally, deep learning models’ dependency high-quality annotation data further constrains their performance improvement. To address these challenges, this study proposes MultiGRNFormer, a Transformer-based model for multi-omics GRN inference. The key innovations of this model include: (1) Integration of transcriptomic and chromatin accessibility data—leveraging a Transformer encoder to effectively capture gene regulatory relationships and improve inference accuracy. (2) Incorporation of a positional encoding mechanism, enabling the model to be sensitive to the order of input features, and meanwhile the use of data augmentation strategies to generate diverse samples, thereby enhancing the utilization of training data. To evaluate the model’s performance, we conducted experiments on seven different single-cell multi-omics datasets and compared it with existing GRN inference methods. The results demonstrate that by combining multi-omics data integration with data augmentation strategies, superior performance in GRN inference tasks can be achieved. Our findings provide new insights for future deep learning-based GRN inference research.