matrix_ls_asc_permutation: ascQTL with permutation in matrix form

In liangyy/mixqtl: Combine total and allele-specific counts for cis-QTL analysis

Description

Perform ascQTL in matrix form with outcome permutated K times

Usage

matrix_ls_asc_permutation(
  asc1,
  asc2,
  x,
  permutation_idx,
  asc_cutoff = 5,
  weight_cap = 100,
  asc_cap = 5000
)

Arguments

asc1	allele-specific read count for haplotype 1 (dimension = N x 1)
asc2	allele-specific read count for haplotype 2 (dimension = N x 1)
x	'genotype' used in ascQTL (which is defined as x1 - x2) (dimension = N x P)
permutation_idx	permutation of samples in each column (dimension = N x K)
asc_cutoff	allele-specific read count cutoff to exclude observations with asc1 or asc2 lower than asc_cutoff
weight_cap	the maximum weight difference (in fold) is min(weight_cap, floor(sample_size / 10)). The ones exceeding the cutoff is capped.
asc_cap	exclude observations with asc1 or asc2 higher than asc_cap

Value

a list of summary statistics beta_hat: estimated log aFC (dimension = K x P, with kth row corresponding to the result under permutation defined in the kth column of permutation_idx) beta_se: standard deviation of log aFC (dimension = K x P) sample_size: number of observations used in regression

Examples

matrix_ls_asc_permutation(
  asc1 = rpois(100, 100),
  asc2 = rpois(100, 100),
  x = matrix(sample(c(0, 0.5, 1), 200, replace = TRUE), ncol = 2),
  permutation_idx = matrix(c(sample(1:100), sample(1:100), sample(1:100)), ncol = 3, byrow = FALSE),
  asc_cutoff = 20,
  weight_cap = 10,
  asc_cap = 1000
)

liangyy/mixqtl documentation built on Sept. 17, 2020, 11:36 a.m.