相关实验视频
Updated: Jun 14, 2026
08:59
Simple, Affordable, and Modular Patterning of Cells using DNA
Published on: February 24, 2021
4.6K
海姆达尔:在单细胞基础模型中进行代币化的模块化框架
Ellie Haber1, Shahul Alam2, Nicholas Ho2
1Machine Learning Department, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA.
bioRxiv : the preprint server for biology
|November 26, 2025
概括
海姆道尔系统地评估了单细胞基础模型 (scFMs) 的代币化策略,揭示了基因身份和顺序对于概括至关重要,特别是在单细胞RNA测序分析中的分布转移下.
科学领域:
- 计算生物学 计算生物学
- 基因组学就是基因组学.
- 机器学习 机器学习
背景情况:
- 基础模型越来越多地用于单细胞RNA测序 (scRNA-seq) 数据分析.
- 这些模型的性能严重依赖于细胞代币化策略,这些策略尚未得到充分理解.
- 开发有效的代币化方法对于推进scRNA-seq分析至关重要.
研究的目的:
- 介绍Heimdall,一个框架和工具包,用于评估单细胞基础模型 (scFMs) 中的代币化策略.
- 系统地评估不同代币化组件对模型性能的影响.
- 提供下一代scFMs可复制的研究和开发的基础.
主要方法:
- 将scFM分解为模块化组件:基因身份编码器 (F_G),表达编码器 (F_E) 和细胞句子构造器 (F_C).
- 使用从头开始训练的变压器模型评估代币化策略.
- 在具有挑战性的转移学习场景中进行评估:跨组织,跨物种和空间基因面板转移,以及反向扰动预测.
主要成果:
- 代币化选择对分发数据的影响最小,但在分发转变时至关重要.
- 基因身份 (F_G) 和令牌顺序显著改善了概括.
- 表达式编码器 (F_E) 提供了额外的性能增长.
- 重组现有策略可以增强模型的概括性.
结论:
- 海姆道尔为评估单细胞代币化策略提供了一种标准化的方法.
- 代币化设计是scFM性能的一个关键决定因素,特别是在分发之外的环境中.
- 海姆道尔公司加速开发更强大,更可通用的scFMs,用于单细胞分析.
相关概念视频
02:00Molecular Models
Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.
01:18Assembly of Cytoskeletal Filaments
Cytoskeletal filaments are polymeric forms of smaller protein subunits. However, individual cytoskeletal filaments may easily disassemble or associate with other similar filaments to form rigid structures. Microfilaments, made of actin monomers, rely on actin-binding proteins to form bundles and create networks of individual actin filaments. Microtubules rely on microtubule-associated proteins (MAPs) to form sturdy cylindrical structures. However, the proteins involved in forming complex...
01:32Assembly of Complex Microtubule Structures
Complex microtubule structures are present in resting cells and in dividing cells. In resting cells, they are responsible for maintaining the cellular architecture, tracks for intracellular transport, positioning of organelles, assembly of cilia and flagella. They mediate the bipolar spindle assembly for chromosomal segregation and positioning of the cell division plate in dividing cells. The formation of microtubule complex structures depends on the cell type, cell stage, and cell function.
01:03Basal Lamina are the Specialized Form of ECM
The basal lamina is a thin extracellular layer that lies underneath the cells and separates them from other tissues. The three layers of the basal lamina are lamina lucida, lamina densa and lamina reticularis. The basal lamina, a mixture of glycoproteins and collagen, provides an attachment site for the epithelium, separating it from underlying connective tissue. The framework of basal lamina has other essential proteins such as laminins mesh, perlecan, entactin, and type IV collagen.
Proteins...
Proteins...
01:21Mechanistic Models: Overview of Compartment Models
Mechanistic models, a category encompassing both physiological and compartmental modeling, differ from empirical models' approaches to incorporating known factors about the systems being modeled. Empirical models describe data with minimal assumptions, while mechanistic models aim to provide a robust description of available data by specifying assumptions and integrating known factors about the system. Compartmental analysis is a key example of a mechanistic model in pharmacokinetics and...
01:23Mechanistic Models: Compartment Models in Individual and Population Analysis
Mechanistic models are utilized in individual analysis using single-source data, but imperfections arise due to data collection errors, preventing perfect prediction of observed data. The mathematical equation involves known values (Xi), observed concentrations (Ci), measurement errors (εi), model parameters (ϕj), and the related function (ƒi) for i number of values. Different least-squares metrics quantify differences between predicted and observed values. The ordinary least squares (OLS)...