TestSingleRegion: Test associations between phenotype and RNA editing levels.

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

Description

Test associations between phenotype and RNA editing levels in a single site or summarized RNA editing levels in a single region.

Usage

TestSingleRegion(
  rnaEdit_num,
  modelPrep_ls,
  respType = c("binary", "continuous", "survival")
)

Arguments

rnaEdit_num	A named numeric vector of (summarized) RNA editing level values with sample IDs as names.
modelPrep_ls	A list includes modelFormula_char which is created by function MakeModelFormula, pheno_df which is the input phenotype data frame in TestAssociations, and minSize (minimum sample size per group to use regular logistic regression) which is created by function CountSamplesPerGroup when respType is "binary".
respType	Type of outcome. Defaults to "binary".

Details

minSize is used by function TestSingleRegion to decide on whether to use regular logistic regression or Firth corrected logistic regression ("https://www.jstor.org/stable/2336755").

Value

a dataframe with test statistics (estimate, stdErr, pValue or coef, exp_coef, se_coef, pValue).

Examples

  data(rnaedit_df)
  
  exm_pheno <- readRDS(
    system.file(
    "extdata",
    "pheno_df.RDS",
    package = 'rnaEditr',
    mustWork = TRUE
    )
  )
  
  exm_model <- list(
    modelFormula_char = "age_at_diagnosis ~ rnaEditSummary",
    pheno_df = exm_pheno,
    minSize = NULL
  )
  
  TestSingleRegion(
    rnaEdit_num = unlist(rnaedit_df[2,]),
    modelPrep_ls = exm_model,
    respType = "continuous"
  )

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