HyCLoST：用于空间转录组学的双曲对比学习与蕴含
原标题：Hyperbolic Contrastive Learning with Entailment for Spatial Transcriptomics
AI 摘要
研究团队提出 HyCLoST，一种用于空间转录组学的双曲对比学习框架，通过双曲几何与基因到图像的蕴含损失，捕捉基因表达与组织形态之间的层级和非对称关系。该方法基于 UNI2-h 与 Geneformer 预训练模型提取特征，并在 SpaRED 基准的 26 个数据集上评估，相比此前方法平均 MSE 降低 6%、PCC 提升 8%。源代码已公开。
正文节选
Hyperbolic Contrastive Learning with Entailment for Spatial Transcriptomics Abstract Spatial Transcriptomics (ST) has transformed biomedical research by enabling the spatial mapping of gene expression across tissue sections. However, high operational costs, specialized equipment requirements, and sensitivity to experimental noise limit the accessibility and scalability of ST. Recent computer vision approaches aim to overcome these limitations by predicting spatial gene expression directly from h