细胞FM:一个大规模的基础模型,预先训练了1亿个人类细胞的转录组学
Yuansong Zeng1,2,3, Jiancong Xie4, Ningyuan Shangguan4
1School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China. zengys@cqu.edu.cn.
Nature communications
|May 20, 2025
概括
我们开发了CellFM,这是一个大规模的单细胞基础模型,在1亿个人类细胞上进行了训练. 细胞FM增强了细胞注释,扰乱预测和基因功能分析,克服了现有模型的局限性.
科学领域:
- 基因组学就是基因组学.
- 计算生物学 计算生物学
- 生物信息学是一种生物信息学.
背景情况:
- 单细胞测序提供高分辨率的转录组数据,揭示细胞异质性.
- 目前的单细胞数据分析面临着诸如噪音,批量效应和稀疏性等挑战.
- 现有的人类单细胞基础模型受到数据大小和参数数量的限制.
研究的目的:
- 开发一个强大的单细胞基础模型 (CellFM),能够准确地表示细胞状态.
- 在数据大小和参数方面解决当前人类基础模型的局限性.
- 改进单细胞转录组数据的分析.
主要方法:
- 收集了包含1亿个人类细胞的多样化数据集.
- 训练了一种具有8亿参数的单细胞基础模型 (CellFM).
- 在MindSpore平台上使用修改后的RetNet框架进行高效的培训.
主要成果:
- 与现有模型相比,CellFM表现出卓越的性能.
- 在单元格注释方面取得了最先进的结果.
- 在扰乱预测,基因功能预测和基因与基因关系捕获方面展示了改进的能力.
结论:
- 细胞FM代表了单细胞数据分析的重大进步.
- 该模型有效地克服了噪音,批量效应和单细胞数据的稀疏性.
- 细胞FM为了解细胞异质性和生物机制提供了一个强大的工具.
更多相关视频
相关概念视频
01:58Cell Specific Gene Expression
Multicellular organisms contain a variety of structurally and functionally distinct cell types, but the DNA in all the cells originated from the same parent cells. The differences in the cells can be attributed to the differential gene expression. Liver cells, whose functions include detoxification of blood, production of bile to metabolize fats, and synthesis of proteins essential for metabolism, must express a specific set of genes to perform their functions. Gene expression also varies with...
01:58Cell Specific Gene Expression
Multicellular organisms contain a variety of structurally and functionally distinct cell types, but the DNA in all the cells originated from the same parent cells. The differences in the cells can be attributed to the differential gene expression. Liver cells, whose functions include detoxification of blood, production of bile to metabolize fats, and synthesis of proteins essential for metabolism, must express a specific set of genes to perform their functions. Gene expression also varies with...
01:16Cell Lines
A cell line is a population of cells grown in vitro that can be subcultured over several generations. Normal cells cease to divide after a certain number of cell divisions, a process known as replicative senescence. This number, called the Hayflick limit, was conceptualized by Leonard Hayflick in 1961 when he observed that fetal cells grown in culture could only divide 40-60 times. This limit is due to the shortening of the telomeres during each round of cell division, preventing cell division...