R package for handling cellatac pipeline output
devtools::install_github("emdann/cellatacUtils")Load pipeline output as Seurat object
library(cellatacUtils)
results.dir <- "path/to/results"
cellatac.seu <- load_cellatac_seurat(results.dir)Calculate QC metrics for peaks
## Load Ensembl based annotation from Bioconductor
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("EnsDb.Hsapiens.v86")
library(EnsDb.Hsapiens.v86)
## Load Signac for ENCODE blacklist annotation
library(Signac)
cellatac.seu <- compute_peakQC(cellatac.seu,
cluster_col = "seurat_clusters",
EnsDb.genome = EnsDb.Hsapiens.v86,
blacklist_gr = blacklist_hg38)Calculated QC metrics:
- tot_count: total count of fragments overlapping peak
- tot_cells: total number of cells in which peak is accessible
- frac_cells: fraction of cells in which peak is accessible
- max_frac: maximum percent of cells with accessible peak in any given cluster (to filter out peaks that are not observed to be accessible in at least n% cells of cells in at least one cluster)
- peak_width: width of peak in bps
- exon: is peak overlapping exon?
- intron: is peak overlapping intron?
- promoter: is peak overlapping promoter?
- annotation: annotation of overlapping genomic region (promoter, exon, intron, intergenic)
- gene_name: name of overlapping gene (downstream gene for promoter region, NA for intergenic peaks)
- gene_id: ENSEMBL id of overlapping gene (downstream gene for promoter region, NA for intergenic peaks)
- tss_distance: distance to closest Transcription Start Site in bps
- ENCODE_blacklist: is peak overlapping ENCODE blacklist region?