Enhanced Integration of Single-Cell Multi-Omics Data Using Graph Attention Networks

作     者：Liao, Xingyu Li, Yanyan Li, Shuangyi Wen, Long Li, Xingyi Yu, Bin 

作者机构：Northwestern Polytech Univ Sch Comp Sci Xian 710129 Shaanxi Peoples R China Qingdao Univ Sci & Technol Sch Data Sci Qingdao 266061 Peoples R China 

出 版 物：《ACS SYNTHETIC BIOLOGY》 (ACS Synth. Biol.)

年 卷 期：2025年第14卷第3期

页      面：931-942页

核心收录：

基　　金：National Natural Science Foundation of China [62172248, 61932018] Natural Science Foundation of Shandong Province of China [ZR2021MF098] King Abdullah University of Science and Technology [FCC/1/1976-44-01, FCC/1/1976-45-01, URF/1/4379-01-47201, REI/1/4742-01-01] 

主　　题：single-cell multiomics integration analysis autoencoder network multihead graph attention mechanism high dimensionality cell type annotation 

摘      要：The continuous advancement of single-cell multimodal omics (scMulti-omics) technologies offers unprecedented opportunities to measure various modalities, including RNA expression, protein abundance, gene perturbation, DNA methylation, and chromatin accessibility at single-cell resolution. These advances hold significant potential for breakthroughs by integrating diverse omics modalities. However, the data generated from different omics layers often face challenges due to high dimensionality, heterogeneity, and sparsity, which can adversely impact the accuracy and efficiency of data integration analyses. To address these challenges, we propose a high-precision analysis method called scMGAT (single-cell multiomics data analysis based on multihead graph attention networks). This method effectively coordinates reliable information across multiomics data sets using a multihead attention mechanism, allowing for better management of the heterogeneous characteristics inherent in scMulti-omics data. We evaluated scMGAT s performance on eight sets of real scMulti-omics data, including samples from both human and mouse. The experimental results demonstrate that scMGAT significantly enhances the quality of multiomics data and improves the accuracy of cell-type annotation compared to state-of-the-art methods. scMGAT is now freely accessible at https://***/Xingyu-Liao/scMGAT.