作为标记的细胞:语言模型中的高维几何和细胞嵌入
1Department of Physics, The University of Texas at Austin, Austin, TX 78712, United States.
Bioinformatics (Oxford, England)
|October 30, 2025
概括
这次审查将自然语言处理 (NLP) 与单细胞分析联系起来. 在NLP基础模型的进步可以通过增强数据表示来改善细胞地图和虚拟细胞模型.
科学领域:
- 计算生物学 计算生物学
- 生物信息学是一种生物信息学.
- 基因组学就是基因组学.
背景情况:
- 单细胞测序产生了代表细胞活动的高维数据.
- 虚拟细胞模型通过从细胞图谱中预先训练的模式来丰富细胞表示.
- 自然语言处理 (NLP) 和单细胞分析共享数据处理范式.
研究的目的:
- 探索NLP嵌入结构如何为单细胞数据分析提供信息.
- 突出NLP基础模型对细胞图谱和虚拟细胞模型的应用.
- 将NLP的进步与强大的细胞模型的开发联系起来.
主要方法:
- 在NLP和单细胞生物学交集的文献审查.
- 分析数据标记化和高维嵌入空间中的并行.
- 检查令牌上下文和低维多元体对数据解释的影响.
主要成果:
- 无论是NLP还是单细胞数据,都涉及将信息分成矢量空间内的代码.
- 符号上下文显著影响嵌入空间几何在两个领域.
- 低维分组对于嵌入空间的稳定性和可解释性至关重要.
- NLP基础模型的发展,如解释性探测和上下文推理,提供了有价值的见解.
结论:
- NLP技术,特别是基础模型,可以增强细胞图谱的构建.
- 虚拟细胞模型可以从NLP启发的方法中获益,以改善表示学习.
- 来自NLP的跨学科见解可以促进单细胞数据分析和解释.
相关概念视频
Cell Size
124.7K
Cell sizes vary widely among and within organisms. Bacterial cells range between 1-10 micrometers (μm)and are considerably smaller than most eukaryotic cells. The smallest bacteria are 0.1 μm in diameter—about a thousand times smaller than eukaryotic cells, which typically range from 10-100 μm.
Surface Area
Cells can take in nutrients and water via diffusion through the plasma membrane itself or through specific channels in the membrane. The area of the membrane surrounding...
Surface Area
Cells can take in nutrients and water via diffusion through the plasma membrane itself or through specific channels in the membrane. The area of the membrane surrounding...
124.7K
Cell Diversity
4.6K
The concept of a cell started with microscopic observations of dead cork tissue by Robert Hooke in 1665. Hooke coined the term "cell" based on the resemblance of the small subdivisions in the cork to the rooms that monks inhabited, called cells. About ten years later, Antonie van Leeuwenhoek became the first person to observe the living and moving cells under a microscope. In the century that followed, the theory that cells represented the basic unit of life developed.
Multicellular...
Multicellular...
4.6K
Cell-surface Signaling
53.7K
Hormones—or any molecule that binds to a receptor, known as a ligand—that are lipid-insoluble (water-soluble) are not able to diffuse across the cell membrane. In order to be able to affect a cell without entering it, these hormones bind to receptors on the cell membrane. When a first messenger, a hormone, binds to a receptor, a signal cascade is set off, causing second messengers, proteins inside the cell, to become activated, resulting in downstream effects.
53.7K
Cell Migration
18.4K
Cell migration, the process by which cells move from one location to another, is essential for the proper development and viability of organisms throughout their life. When cells are not able to migrate properly to their ordained locations, various disorders may occur. For example, disruption in cell migration causes chronic inflammatory diseases such as arthritis.
18.4K
Cell Migration
6.3K
Cell migration is a process by which the cells move from one location to another, playing an essential role in embryological development, repair and regeneration, immune response, and metastasis. Cells migrate in response to chemical or mechanical signals generated by specific organs or tissues. The overall mechanism includes three steps - polarization, protrusion, and release. Polarization involves the formation of a distinct cell front and rear, which determines the direction of movement.
6.3K
Cellular Differentiation
4.9K
How does a complex organism such as a human develop from a single cell? It all starts from a single fertilized egg which gives rise to a vast array of cell types, such as nerve cells, muscle cells, and epithelial cells that characterize the adult? Throughout development and adulthood, cellular differentiation leads cells to assume their final morphology and physiology. Differentiation is the process by which unspecialized cells become specialized to carry out distinct functions.
A zygote is a...
A zygote is a...
4.9K


