runFigRGRN: Infer FigR TF-DORC associations

In buenrostrolab/FigR: Functional Inference of Gene Regulation (FigR) using paired single cell ATAC-seq and RNA-seq data

Infer FigR TF-DORC associations

Description

Function to run TF motif-to-gene associations using reference DORC peak-gene mappings and TF RNA expression levels

Usage

runFigRGRN(
  ATAC.se,
  dorcK = 30,
  dorcTab,
  n_bg = 50,
  genome,
  dorcMat,
  rnaMat,
  dorcGenes = NULL,
  nCores = 1
)

Arguments

ATAC.se	SummarizedExperiment object of peak x cell scATAC-seq data, the same as used to compute DORCs using runGenePeakcorr
dorcK	numeric specifying the number of dorc nearest-neighbors to pool peaks from for the motif enrichment per DORC. Default is 30, i.e. set to ~3 percent of total DORCs determined
dorcTab	data.frame object containing significant peak-gene pairs using which DORC scores will be computed. Must be a filtered set returned from runGenePeakcorr. IMPORTANT: Make sure the exact same scATAC SE peak set was used when determining DORCs that is used here to get corresponding DORC peak counts
n_bg	number of background peaks to use for
genome	character specifying a valid genome assembly to use for peak GC content estimation and background peak determination. Must be one of "hg19","hg38", or "mm10", and requires the corresponding genomes package e.g. BSgenome.Hsapiens.UCSC.hg19 for hg19
dorcMat	Matrix object of smoothed single-cell DORC accessibility scores
rnaMat	Matrix object of smoothed single-cell RNA expression values
dorcGenes	character vector specifying the subset of DORCs to test, if not running on everything. Note: We still use the entire list of DORCs found in dorcMat to determine dorc KNNs from, but will only test and include results for these specified genes (also must exist in the provided RNA matrix rnaMat as rownames)
nCores	numeric specifying the number of cores to run DORCs in parallel. Default is 1, i.e. don't use parallel backend

Value

a data.frame with all TF-DORC motif enrichment and correlation associations, and the corresponding FigR regulation score for each association

Author(s)

Vinay Kartha

buenrostrolab/FigR documentation built on Feb. 10, 2023, 11:26 a.m.