This repository is a resource to demonstrate how the robin (Robustness in Network) package (>=v.2) can be applied to single cell data, in order to answer the following question:
How well do clusters approximate cell subtypes?
By exploring the Articles present at this website https://drighelli.github.io/scrobinv2/, we show how the made improvements to the robin package helps to speed up the computations in order to apply the implemented models to high-dimentional single cell data.
Robin gives the possibility to choose the community detection algorithm that better fits your single cell network and to investigate if the communities that it gives are statistically significant.
Package is at: https://cran.r-project.org/package=robin
Citations:
- Policastro, V., Righelli, D., Carissimo, A., Cutillo, L. & De Feis, I. ROBustness In Network (robin): an R package for Comparison and Validation of communities. The R Journal 13, 292–309. doi:10. 32614/rj-2021-040 - Link to the paper