{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"萌喵读文献-生物信息学","title":"今日生物信息学最高分文献 - 2025-09-26","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/37693f08\"></iframe>","width":"100%","height":180,"duration":310,"description":"科研喵使用ai读文献，祝你效率百倍，访问labcat.com.cn下载。本期关注发表在Nature Methods（IF=36.1）上的重要研究\"EpiAgent: foundation model for single-cell epigenomics\"。该研究 introduces EpiAgent，一种在大型人类scATAC-Corpus上预训练的单细胞表观基因组学基础模型。EpiAgent能将染色质可及性模式编码为\"细胞句子\"，通过双向注意力捕获细胞异质性。研究显示，EpiAgent在无监督特征提取、细胞类型注释和数据插补等任务中表现出色，并能预测细胞响应、整合数据并模拟细胞状态变化。这项突破性工具使研究人员能够以零样本方式直接注释细胞类型，为理解复杂疾病机制和开发精准疗法提供了强大新工具。","thumbnail_url":"https://img.transistorcdn.com/38WU46uKrju37cyiLrKcVSktFL4vxBb_oTgbpM_2CRw/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS8zOGU2/ZmEyYWE0MDZkNGFj/NWMzNGY5ZmU4YTk0/ZTBlNS5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}