Skip to content

An R package for visualization and relational analysis of cell states, based on the Mapper algorithm from topological data analysis.

Notifications You must be signed in to change notification settings

matei-ionita/HiTMapper

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

51 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

HiTMapper

Installation

It's easiest to install HiTMapper with the help of the devtools package.

install.packages("devtools") # if necessary
devtools::install_github("matei-ionita/HiTMapper")

Summary

Mapper is an algorithm developed for topological data analysis, originally due to [1]. It constructs a graph which approximates the (hypothetical) landscape from which data points are sampled. HiTMapper is an implementation which is tailored to cytometry data: for example, only using subroutines that run in linear time, so that millions of data points can be processed in a few minutes.

Each node in the Mapper graph represents multiple data points, resulting in a compact visualization of the dataset. Crucially, the nodes are fine-grained, not meant to represent distinct phenotypes, but rather subtly different cell states. This enables the visualization of hundreds of detailed cell states in a relational framework. We can also use the graph to provide a warm start for traditional clustering, using the Leiden method[2] for community detection.

visualization

References

[1] Singh, Gurjeet, Facundo Mémoli, and Gunnar E. Carlsson. "Topological methods for the analysis of high dimensional data sets and 3d object recognition." SPBG 91 (2007): 100.

[2] Traag, Vincent A., Ludo Waltman, and Nees Jan Van Eck. "From Louvain to Leiden: guaranteeing well-connected communities." Scientific reports 9.1 (2019): 1-12.

About

An R package for visualization and relational analysis of cell states, based on the Mapper algorithm from topological data analysis.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published