burden_test: Carry out a Fisher's exact test

In adamwaring/ClusterBurden: Rare variant association utilising burden and positional information.

Description

Burden test with formatting step to make a contingency table

Usage

burden_test(
  n1 = NULL,
  n2 = NULL,
  ss1 = NULL,
  ss2 = NULL,
  pval = T,
  case_coverage = NULL,
  control_coverage = NULL,
  cov_threshold = 0.5,
  alternative = "greater"
)

Arguments

n1	number of carriers cases
n2	number of carriers controls
ss1	sample size cases
ss2	sample size controls
pval	return p-value or full output?
case_coverage	optional coverage data for cases in format: data.table(symbol, protein_position, over_10)
control_coverage	optional coverage data for controls in format: data.table(symbol, protein_position, over_10)
cov_threshold	threshold at which to exclude a residue position from the analysis (choose 0 to keep all residues)
alternative	indicates the alternative hypothesis and must be one of "two.sided", "greater" or "less".

Details

Fisher.test from base R with extra formatting steps.

For a one-sided test i.e. excess in cases, use the default alternative="greater"

if pval=F Use $p for p-value Use $est for odds-ratio Use $conf.int[1:2] for 95

Value

Returns an object of class htest - use $p for p-value

Author(s)

Adam Waring - [e-mail]

Examples

# For the simple case where you have the vector status (binary case or control) and a vector of whether they
have a variant 'car' (presumably after filtering) then;

N = 100
car = sample(c(0, 1), N, rep=T)
status = sample(c(0, 1), N, rep=T)

burden_test(car, status)

burden_test(n1=10, n2=20, ss1=100, ss2=100)

adamwaring/ClusterBurden documentation built on July 29, 2020, 9:50 p.m.