返回
An unsupervised method for spatial transcriptomics analysis based on adversarial autoencoder
DOI:10.1093/bib/bbag070.png)
摘要
En 中文
空间转录组学(ST)为解码基因表达的空间组织提供了前所未有的机会,但ST数据固有的噪声和复杂性给准确分析带来了巨大挑战。本文提出DACN,一个统一框架,将改进的对抗自编码器(AAE)与图卷积网络(GCN)相结合,以稳健地分析不同分辨率和通量下的ST数据。DACN采用混合编码器,结合多头注意力与残差连接,捕捉精细局部表达模式的同时保留关键全局信息。混合编码器与生成器共同构建AAE模块,用于去噪表达谱并学习稳定的潜在表征。GCN组件进一步利用空间邻域关系优化这些嵌入。在多个不同分辨率的ST数据集中,DACN在准确性和稳健性方面均显著优于现有方法。所有代码和数据集均可通过https://github.com/lanbiolab/DACN公开获取。
Keyword:
spatial transcriptomics
adversarial autoencoder
deep learning
spatial domain
期刊
IF:
7.7
论文数:
5.8K
被引数:
2.7W
机构
引用论文
Spotiphy enables single-cell spatial whole transcriptomics across an entire sectionSpotiphy 实现了整个切片的单细胞空间全转录组学。
NATURE METHODS
IF32.1
Spatially resolved cell atlas of the mouse primary motor cortex by MERFISHMERFISH对小鼠初级运动皮层的空间分辨细胞图谱
Nature
IF0
SiGra: single-cell spatial elucidation through an image-augmented graph transformerSiGra: 通过图像增强的图形转换器进行单细胞空间阐释
NATURE COMMUNICATIONS
IF15.7
COO-DuDo: computation overhead optimization methods for dual-domain sparse-view CT reconstructionCOO-DuDo:双域稀疏视图CT重建的计算开销优化方法
Deciphering cell-cell communication at single-cell resolution for spatial transcriptomics with subgraph-based graph attention network
NATURE COMMUNICATIONS
IF15.7
Transcriptome-scale super-resolved imaging in tissues by RNA seqFISH+通过RNA seqFISH在组织中进行转录组规模的超分辨成像
Nature
IF0
Dft\_anpd: a dual-feature two-sided attention network for anticancer natural products detectionDft\_anpd:一种用于抗癌天然产物检测的双特征双向注意力网络
Slide-seq: A scalable technology for measuring genome-wide expression at high spatial resolutionSlide-seq: 一种用于以高空间分辨率测量全基因组表达的可扩展技术
Science
IF0

