regActivityAroundTSS: Identify regulatory regions around provided TSSes

Description Usage Arguments Details Value Examples

Description

The function identifies the regulatory regions around provided TSSes over a pre-defined window. The function needs a GRanges object for TSSes with meta-columns corresponding to expression levels in different cell types or conditions.

Usage

1
2
regActivityAroundTSS(regActivity, geneExpression, upstream = 5e+05,
  downstream = 5e+05, mc.cores = 1, force = FALSE)

Arguments

regActivity

a GRanges object output from regActivity function

geneExpression

a GRanges object output from targetneExp that contains the TSS location and associated gene expression values per cell type or condition as meta data. Each row should have a "name" and "name2" columns for unique id or name/symbol for the gene which the TSS is associated with. One is Ensembl id and the other could be used for the gene symbol. Other metadata column names should represent sample names/ids and should match the GRanges object provided via regActivity argument.

upstream

number of basepairs upstream from TSS to look for regulatory regions. default 500kb

downstream

number of basepairs downstream from TSS to look for regulatory regions. default 500kb

mc.cores

(def:1) Define the number of cores to use; at most how many child processes will be run simultaneously using mclapply from parallel package. Parallelization requires at least two cores.

force

(DEFAULT: FALSE) This argument allows to test input enhancers and genes in 1-to-1 manner. Meaning,a gene expression of gene1 is modelled using enhancer1 enhancer activity, gene2 gene expression is modelled with enhancer2 enhancer activity, etc. Thus, input gene expression GRanges object (geneExpression) needs to have equal length as enhancer activity GRanges object (regActivity). It overwrites upstream and downstream arguments since 1-to-1 relationship is tested.

Details

only enhancers located within (+/-)upstream/downstream of TSS are identified,extracted and reported in output (together with info about gene expression). Sample id's (corresponding to the cell types or conditions) are included only if both, 1) gene expression values and 2) quantified regulatory activity are available in TSS and regActivity objects. Non-overlapping cell types are excluded.

Value

An output of regActivityAroundTSS() is a GRangesList object that contains per gene GRanges with location of the corresponding TSS and regulatory regions identified around that gene. Names for the GRangesList are unique gene ids/names. Metadata for each GRanges object in the GRangesList represents regulatory activity and gene expression quantified across a number of samples (.bw files) that have matched IDs in RNA-Seq and CHiP-Seq,DNase-Seq or bisulfite sequencing experiment. The GRanges objects have the following metadata columns: 1. featureType: either "gene" or "regulatory" 2. name: name/id for gene and enhancers. Gene name could be id from a database enhancer name should be in the format as follows "chr:start-end" 3. name2: a secondary name for the feature, such as gene symbol "PAX6" etc. not necessary for enhancers could be NA 4. other columns: numeric values for gene expression or regulatory actvity. Column names represent sample names/ids.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
 # INPUT1: CREATING Toy example for quantified gene expression:

geneExpression <- GRReg1_toy
  geneExpression$bw1 <- rep(1,length(GRReg1_toy))
  geneExpression$bw2 <- rep(2,length(GRReg1_toy))
  geneExpression$bw3 <- rep(3,length(GRReg1_toy))
  
  # INPUT2: CREATING TOY example for quantified enhancer activity:
  
regRegion <- GRReg2_toy
   regRegion$bw1 <- rep(3,length(regRegion))
   regRegion$bw2 <- rep(4,length(regRegion))

# Overlapping quantified enhancer activity and gene expression:
# OUTPUT  regActivityAroundTSS():
regActivityAroundTSS(regActivity=regRegion, geneExpression=geneExpression,
upstream=1,downstream=1)

# when different upstream/downstream argument is used:

regActivityAroundTSS(regRegion,geneExpression,upstream=5,downstream=5)

# force=T, forcing 1-to-1 relationship
 regActivityAroundTSS(regRegion[1:length(geneExpression)],
 geneExpression,upstream=5,downstream=5,force=TRUE)

BIMSBbioinfo/reg2gene documentation built on May 3, 2019, 6:42 p.m.