.png){kind=icon}
.png){kind=icon}
Return
BFAST: joint dimension reduction and spatial clustering with Bayesian factor analysis for zero-inflated spatial transcriptomics data
.png){kind=icon}
OA
AI
DOI：10.1093/bib/bbae594.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The development of spatially resolved transcriptomics (ST) technologies has made it possible to measure gene expression profiles coupled with cellular spatial context and assist biologists in comprehensively characterizing cellular phenotype heterogeneity and tissue microenvironment. Spatial clustering is vital for biological downstream analysis. However, due to high noise and dropout events, clustering spatial transcriptomics data poses numerous challenges due to the lack of effective algorithms. Here we develop a novel method, jointly performing dimension reduction and spatial clustering with Bayesian Factor Analysis for zero-inflated Spatial Transcriptomics data (BFAST). BFAST has showcased exceptional performance on simulation data and real spatial transcriptomics datasets, as proven by benchmarking against currently available methods. It effectively extracts more biologically informative low- dimensional features compared to traditional dimensionality reduction approaches, thereby enhancing the accuracy and precision of clustering.
Keywords:
spatial Transcriptomics
Bayesian factor analysis
zero-inflated
dimension reduction
spatial clustering
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.7
Papers:
5.6K
Citations:
2.7W