The performance of Federated Harmony on three types of single cell data.

The performance of Federated Harmony on three types of single cell data.

<p>Each row represents results of a type of data. scRNA-seq data (first row); mouse brain tissue spatial transcriptomics data (second row); scATAC-seq data (third row). For <b>a-c</b>, the upper plot is the iLISI density plot, the lower one is the UMAP. <b>a</b>: iLISI...

Full description

Saved in:
Bibliographic Details
Main Author: Ruizhi Yuan (16864629) (author)
Other Authors: Ziqi Rong (22339789) (author), Haoran Hu (6032510) (author), Tianhao Liu (4678147) (author), Shiyue Tao (11871368) (author), Wei Chen (23863) (author), Lu Tang (655389) (author)
Published: 2025
Subjects:
Microbiology
Genetics
Biotechnology
Infectious Diseases
Plant Biology
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
xlink "> large
based data integration
approach preserves privacy
introduce federated harmony
raising privacy
federated learning
data sources
various types
security concerns
often confounded
harmony algorithm
downstream analysis
distributed multi
cell multi
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:<p>Each row represents results of a type of data. scRNA-seq data (first row); mouse brain tissue spatial transcriptomics data (second row); scATAC-seq data (third row). For <b>a-c</b>, the upper plot is the iLISI density plot, the lower one is the UMAP. <b>a</b>: iLISI density plot and UMAP before integration; <b>b</b>: iLISI density plot and UMAP after Harmony integration; <b>c</b>: iLISI density plot and UMAP after Federated Harmony integration; <b>d</b>: box plots of ARI values of naive k-means clustering results for Harmony-integrated and Federated Harmony-integrated embeddings.</p>

Similar Items