paperKB
coga / coga-kb
Help
Sign in

Chunk #59 — Methods — Cell subtype annotation and testing

Source
Integrated single-cell multiomic profiling of caudate nucleus suggests key mechanisms in alcohol use disorder.
Embedded
yes

Text

Microglia and astrocyte clusters were further divided into four and two subclusters, respectively, by performing another iteration of FindNeighbors and FindClusters, using cells from the 143 samples. Subcluster-specific genes were determined by using the roc test within the FindMarkers function in Seurat (see https://satijalab.org/seurat/reference/findmarkers). The top 50 genes (based on myAUC statistic) were used as input into g:Profiler82 to determine enriched biological pathways specific to that subcluster. (Notably, g:Profiler uses their own method, called gSCS, for multiple testing correction in lieu of Benjamini–Hochberg (BH) method)