inst/www/helpfiles/differential_analysis.md

In vallotlab/ChromSCape: Analysis of single-cell epigenomics datasets with a Shiny App

output: html_document: self_contained: no

Differential Analysis

In this tab, investigate on the differences in levels of the mark of interest or accessibility for scATAC-seq between clusters or samples. Groups can be compared in three kind of ways:

• One vs All: This option will run comparison for each cluster, taking as reference all the cells in other clusters. This is interesting to understand which features are specifically enriched / depleted in one cluster but not any others.

• Pairwise: This option will compare each cluster two by two and consolidate the multiple differential tests for each cluster using scran::combineMarkers function with pval.type = "any" : 'This approach does not explicitly favour [loci] that are uniquely [differential] in a cluster. Rather, it focuses on combinations of [loci] that - together - drive separation of a cluster from the others.

• Custom Using this option, choose your own group and reference to compare, either samples or clusters and do a single differential testing. This is useful when you are only interested in the differences between two groups but not the others.

For each option, user can pick either Wilocoxon non-parametric testing or edgeR GLM parametric testing, relying on linear models.

If running multiple comparisons using various options, all the comparisons will be saved and accessible at any time.

vallotlab/ChromSCape documentation built on Oct. 15, 2023, 1:47 p.m.