R: Run bumphunter

runBumphunter {MEAL}

R Documentation

Run bumphunter

Description

Run bumphunter to a methylation dataset. This function contains all steps of bumphunter analysis, from model.matrix creation to running the analysis.

Usage

runBumphunter(set, model, coefficient = 2, bumphunter_cutoff = 0.1,
  num_permutations = 0, bumps_max = 30000, betas = TRUE,
  check_perms = FALSE, verbose = FALSE, resultSet = FALSE, ...)

Arguments

`set`	`GenomicRatioSet`, `eSet` derived object or `SummarizedExperiment`
`model`	Model matrix or formula to get model matrix from `set`.
`coefficient`	Numeric with the column of model matrix used in the analysis. (Default: 2)
`bumphunter_cutoff`	Numeric with the minimum cutoff to include a probe in a block. (Default: 0.1)
`num_permutations`	Numeric with the number of permutations run to compute the bumps p-value. (Default: 0)
`bumps_max`	Numeric with the maximum number of bumps used in the permutation. This parameter only applies when `num_permutations` is greater than 0. (Default: 30000)
`betas`	If `set` is a `GenomicRatioSet`, should beta values be used? (Default: TRUE)
`check_perms`	Logical. Should we check that there are less bumps than `bumps_max`? This parameter only applies when `num_permutations` is greater than 0. (Default: TRUE)
`verbose`	Logical value. Should the function be verbose? (Default: FALSE)
`resultSet`	Should results be encapsulated in a `resultSet`? (Default: TRUE)
`...`	Further arguments passed to `bumphunter`.

Details

runBumphunter is a wrapper for minfi bumphunter. This function runs all the steps required prior running bumphunter from the methylation set and the formula of the model. This implementation allows running bumphunter to other objects than GenomicRatioSet. The result can be encapsulated in a ResultSet to take adavantege of its plotting capabilities.

If the user wants to run permutations to calculate p-values, this implementation can filter the bumps to avoid doing a very high number of permutations and to reduce computation time. To do so, we can set the maximum number of bumps that we want to permute with the bumps_max parameter. runBumphunter increases bumphunter_cutoff value until the number of bumps is lower than bumps_max.

Value