DisConST: Distribution-aware Contrastive Learning for Spatial Domain Identification

作者：Peimeng Zhen, Xiaofeng Wang, Han Shu, Jialu Hu, Yongtian Wang, Jiajie Peng, Xuequn Shang, Jing Chen, Tao Wang · 发表于：Genomics Proteomics & Bioinformatics · 年份：2025 · DOI：10.1093/gpbjnl/qzaf085 · 被引用次数：7 · 研究领域：Single-cell and spatial transcriptomics、Molecular Biology Techniques and Applications、Domain Adaptation and Few-Shot Learning

Spatial transcriptomics (ST) is a cutting-edge technology that provides comprehensive insights into gene expression (GE) patterns from a spatial perspective. A key research focus within this field is spatial domain identification, which is essential for exploring tissue organization, biological development, and disease mechanisms. Although methods have been developed, they still face challenges in modeling GE information together with spatial locations, resulting in suboptimal accuracy. Here, we introduce Distribution-aware Contrastive Learning for Spatial Transcriptomics (DisConST), a novel deep learning method designed to improve spatial domain detection within ST datasets. DisConST addresses key challenges, such as the high dropout rates and the complex integration of spatial and GE data, by incorporating contrastive learning strategies that are aware of the underlying data distributions. It employs the zero-inflated negative binomial distribution, along with graph contrastive learning, to generate more informative latent representations. These representations efficiently integrate spatial positions, transcriptomic profiles, and cell-type proportions within spots. We validated DisConST across diverse ST datasets of tissues, organs, and embryos from various sequencing platforms in both normal and disease states. Our results consistently demonstrate that DisConST achieves superior spatial domain recognition accuracy compared to existing state-of-the-art methods. Furthermore,...