GetMinPairwiseCor: Calculate minimum pairwise correlation for sub-regions.

In rnaEditr: Statistical analysis of RNA editing sites and hyper-editing regions

Description

Filter the contiguous co-edited subregions found from FindCorrelatedRegions, by calculating pairwise correlations and then selecting subregions passing the minimum correlation filter.

Usage

GetMinPairwiseCor(
  rnaEditCluster_mat,
  minPairCorr = 0.1,
  probes_ls,
  method = c("spearman", "pearson")
)

Arguments

rnaEditCluster_mat	A matrix of RNA editing level values on individual sites, with row names as sample IDs and column names as site IDs in the form of "chrAA:XXXXXXXX".
minPairCorr	Minimum pairwise correlation coefficient of sites within a cluster, used as a filter. To use this filter, set a number between -1 and 1 (defaults to 0.1). To turn it off, please set the number to -1.
probes_ls	A list of regions with sites. Please note that probes in each list need to be ordered by their locations.
method	Method for computing correlation. Defaults to "spearman".

Value

A list with a list of probes passing the minPairCorr and a data frame with the following columns:

• subregion : index for each output contiguous co-edited region.

• keepminPairwiseCor : indicator for contiguous co-edited subregions, The regions with keepminPairwiseCor = 1 passed the minimum correlation and will be returned as a contiguous co-edited subregion.

• minPairwiseCor : the minimum pairwise correlation of sites within a subregion.

Examples

  data(t_rnaedit_df)
  
  ordered_cols <- OrderSitesByLocation(
    sites_char = colnames(t_rnaedit_df),
    output = "vector"
  )
  exm_data <- t_rnaedit_df[, ordered_cols]
  
  exm_sites <- list(
    "1" = c("chr1:28661656", "chr1:28661718", "chr1:28662148")
  )
  
  GetMinPairwiseCor(
    rnaEditCluster_mat = exm_data,
    minPairCorr = 0.1,
    probes_ls = exm_sites,
    method = "spearman"
  )
               

rnaEditr documentation built on Nov. 8, 2020, 8:26 p.m.