gQTLstats (version 1.4.2)

qqStore: create a binned QQplot for a sharded store

Description

create a binned QQplot for a sharded store with association and permutation statistics

Usage

qqStore(st, ids = NULL, 
  .probs = c(0, seq(0.6, 0.8, 0.2), 0.9, 0.95, 0.99, 0.999, 0.9999, 1), 
  xlim.in = c(0.2, 75), lowfac = 0.5, xlab = "Permutation distribution", 
  ylab = "Distribution of score statistic", countpos = 50, 
  plot.it = TRUE, doab = TRUE, scoreField = "chisq", 
  permField = "permScore_1", ...)

Arguments

st

instance of ciseStore-class

ids

optional job id vector; if NULL, all jobs used

.probs

vector of probabilities for use with quantile evaluation, as provided in ffbase, using storeToQuantiles

xlim.in

xlim setting for QQplot

lowfac

we use a log-log plot, and the first quantile (as prescribed in .probs) is often close to zero; we reassign it to lowfac*(second quantile)

xlab

label

ylab

label

countpos

where on the x axis will we stack the information on bin counts

plot.it

logical, if FALSE, a list is returned with elements on quantile values and bin counts

doab

logical prescribing drawing of line of identity

scoreField

tag in store naming the statistic, typically 'chisq', can also be 'tstat' for GTEx

permField

tag in store naming the field holding statistics on realizations from permutation distribution

...

passed to storeToQuantiles

Value

• invisibly returns list with elements qx, qy, counts, fracs

Examples

library(geuvStore2)
library(gQTLBase)
gs = makeGeuvStore2()
qqStore(gs) #, ids=partialIds()[1:20])