eset_presence_absence: Presence/Absence Testing

In vladpetyuk/vp.misc: Miscellaneous functions used by Vlad Petyuk for proteomics data analysis

Description

A function for testing the significance in differences of presence/absence patterns.

Usage

eset_presence_absence(
  eset,
  grouping,
  test = c("fisher", "chisq", "g", "binom"),
  ...
)

Arguments

eset	eset (or most likely eset subclass) object
grouping	character defining the column in phenoData
test	character: Fisher exact test, χ^2, G-test or binomial.
...	other aruments to pass to the test functions (e.g. simulate.p.value = TRUE)

Details

The recommended way of using the test is two-group comparison. All except "binom" options will work for larger number of groups.

Value

data.frame

p.value

p-value

effect

difference between second and first group (according to the factor levels). In case of larger number of levels, it is the absolute value of the max diffence

...

proportions of present for each group. Column names are level names.

Note

see also http://www.ncbi.nlm.nih.gov/pubmed/20831241

Examples

library("MSnbase")
Nsam = 20
Npep = 5
M <- matrix(rbinom(Nsam*Npep, 1, 0.5), ncol=Nsam)
pd <- data.frame(group = gl(2, Nsam/2))
fd <- data.frame(otherfdata = letters[1:Npep])
x0 <- MSnSet(M, fd, pd)
eset_presence_absence(x0, 'group', test='fisher')
eset_presence_absence(x0, 'group', test='chisq', simulate.p.value=TRUE)
eset_presence_absence(x0, 'group', test='g')
eset_presence_absence(x0, 'group', test='binom')

vladpetyuk/vp.misc documentation built on June 25, 2021, 6:35 a.m.