runOnEachSynCHR: Simulate a multigeneration methylation experiment with...
In methInheritSim: Simulating Whole-Genome Inherited Bisulphite Sequencing Data

Description Usage Arguments Value Author(s) Examples

View source: R/methylInheritanceSimInternalMethods.R

Simulate a multigeneration methylation case versus control experiment with inheritance relation using a real control dataset.

runOnEachSynCHR(methData, nbSynCHR, nbSimulation, nbBlock, nbCpG, nbGeneration,
  vNbSample, vpDiff, vpDiffsd, vDiff, vInheritance, rateDiff, minRate,
  propInherite, propHetero, keepDiff, outputDir, fileID, minReads, maxPercReads,
  meanCov, context, assembly, saveGRanges, saveMethylKit, runAnalysis, nbCores,
  vSeed)

`methData`	an object of class `methylBase`, the CpG information from controls (CTRL) that will be used to create the synthetic chromosome. The `methData` object can also contain information from cases but only the controls are used.
`nbSynCHR`	a positive `integer`, the number of distinct synthetic chromosomes that will be generated.
`nbSimulation`	a positive `integer`, the number of simulations generated for each parameter (`vNbSample`, `vpDiff`, `vDiff` and `vInheritance`). The total number of simulation is nbSimulation * `length(vNbSample)` * `length(vpDiff)` * `length(vInheritance)`)
`nbBlock`	a positive `integer`, the number of blocks used for sampling.
`nbCpG`	a positive `integer`, the number of consecutive CpG positions used for sampling from `methInfo`.
`nbGeneration`	a positive `integer`, the number of generations simulated.
`vNbSample`	a `vector` of distinct positive `integer`, the number of controls (CTRL) and cases in the simulated dataset. In the simulated dataset, the number of CTRL equals the number of cases. The number of CTRL do not need to be equal to the number of Case in the real `methData` dataset.
`vpDiff`	a `vector` of distinct `double` superior to `0` and inferior or equal to `1`, the mean value for the proportion of samples that will have, for a specific position, differentially methylated values. It can be interpreted as the penetrance. Note that `vpDiff` and `vpDiffsd` must be the same length.
`vpDiffsd`	a `vector` of a non-negative `double`, the standard deviation associated to the `vpDiff`. Note that `vpDiff` and `vpDiffsd` must be the same length.
`vDiff`	a `vector` of distinct non-negative `double` included in [0,1], the proportion of C/T for a case differentially methylated that follows a beta distribution where the mean is shifted by `vDiff` from the CTRL distribution.
`vInheritance`	a `vector` of distinct non-negative `double` included in [0,1], the proportion of cases that inherits differentially methylated sites.
`rateDiff`	a positive `double` inferior to `1`, the mean of the chance that a site is differentially methylated.
`minRate`	a non-negative `double` inferior to `1`, the minimum rate for differentially methylated sites.
`propInherite`	a non-negative `double` inferior or equal to `1`, the proportion of differentially methylated regions that are inherated.
`propHetero`	a non-negative `double` between [0,1], the reduction of `vDiff` for the second and following generations.
`keepDiff`	a `logical`, when `TRUE`, the differentially methylated sites will be the same for all simulated datasets. Datasets generated using differents parameter values from vector parameters (`vpDiff`, `vDiff` and `vInheritance`) wil all have the same differentially methylated sites.
`outputDir`	a string of `character` or `NULL`, the path where the files created by the function will be saved. When `NULL`, the files are saved in a directory called "outputDir" that is located in the current directory.
`fileID`	a string of `character`, a identifiant that will be included in each output file name. Each output file name is composed of those elements, separated by "_": a type name, ex: methylGR, methylObj, etc.. a `fileID` the chromosome number, a number between 1 and `nbSynCHR` the number of samples, a number in the `vNbSample` `vector` the mean proportion of samples that has, for a specific position, differentially methylated values, a number in the `vpDiff` `vector` the proportion of C/T for a case differentially methylated that follows a shifted beta distribution, a number in the `vDiff` `vector` the proportion of cases that inherits differentially sites, a number in the `vInheritance` `vector` the identifiant for the simulation, a number between 1 and `nbSimulation` the file extension ".rds"
`minReads`	a positive `integer`, sites and regions having lower coverage than this count are discarded. The parameter corresponds to the `lo.count` parameter in the `methylKit` package.
`maxPercReads`	a `double` between [0,100], the percentile of read counts that is going to be used as upper cutoff. Sites and regions having higher coverage than `maxPercReads` are discarded. This parameter is used for both CpG sites and tiles analysis. The parameter correspond to the `hi.perc` parameter in the `methylKit` package.
`meanCov`	a positive `integer`, the mean of the coverage at the simulated CpG sites.
`context`	a string of `character`, the short description of the methylation context, such as "CpG", "CpH", "CHH", etc..
`assembly`	a string of `character`, the short description of the genome assembly, such as "mm9", "hg18", etc..
`saveGRanges`	a `logical`, when `true`, the package save two files type. The first generate for each simulation contains a `list`. The length of the `list` corresponds to the number of generation. The generation are stored in order (first entry = first generation, second entry = second generation, etc..). All samples related to one generations are contained in a `GRangesList`. The `GRangeaList` store a `list` of `GRanges`. Each `GRanges` stores the raw mehylation data of one sample. The second file a numeric `vector` denoting controls and cases (a file is generates by entry in the `vector` parameters `vNbSample`).
`saveMethylKit`	a `logical`, when `TRUE`, for each simulations save a file contains a `list`. The length of the `list` corresponds to the number of generation. The generation are stored in order (first entry = first generation, second entry = second generation, etc..). All samples related to one generations are contained in a S4 `methylRawList` object. The `methylRawList` object contains two Slots: 1. treatment: A numeric `vector` denoting controls and cases. 2. .Data: A `list` of `methylRaw` objects. Each object stores the raw methylation data of one sample.
`runAnalysis`	a `logical`, if `TRUE`, two files are saved for each simulation: 1. The first file is the methylObj... file formated with the `methylkit` package in a S4 `methylBase` object (using the `methylKit` functions: `filterByCoverage`, `normalizeCoverage` and `unite`). 2. The second file contains a S4 `calculateDiffMeth` object generated using the `methylKit` functions `calculateDiffMeth` on the first file.
`nbCores`	a positive `integer`, the number of cores used when creating the simulated datasets. Default: `1` and always `1` for Windows.
`vSeed`	a `integer`, a seed used when reproducible results are needed. When a value inferior or equal to zero is given, a random integer is used. .

0 indicating that the function have been successful.

Pascal Belleau, Astrid Deschenes

## Load dataset containing methyl information
data(samplesForChrSynthetic)

## Set the output directory where files will be created
temp_dir <- "test_runOnEachSynCHR"

## Create directory
if(!dir.exists(temp_dir)) {
    dir.create(temp_dir, showWarnings = TRUE)
}

## Create 2 simulated dataset (nbSimulation = 2) 
## over 3 generations (nbGenration = 3) with
## 6 cases and 6 controls (nNbsample = 6) using only one set
## of parameters (vpDiff = 0.9, vpDiffsd = 0.1, vDiff = 0.8)
methInheritSim:::runOnEachSynCHR(methData = samplesForChrSynthetic, 
    nbSynCHR = 1, nbSimulation = 2, nbBlock = 10, nbCpG = 20, 
    nbGeneration = 3, vNbSample = c(6), vpDiff = c(0.9), vpDiffsd = c(0.1), 
    vDiff = c(0.8), vInheritance = c(0.5), propInherite = 0.3, 
    rateDiff = 0.3, minRate = 0.2, propHetero = 0.5, keepDiff = FALSE, 
    outputDir = temp_dir, fileID = "F1",  minReads = 10, 
    maxPercReads = 99.9, meanCov = 80, context = "CpG", assembly="Rnor_5.0",
    saveGRanges = FALSE, saveMethylKit = FALSE,
    runAnalysis = FALSE, nbCores = 1, vSeed = 32)

## Delete the output directory and its content
if (dir.exists(temp_dir)) {
    unlink(temp_dir, recursive = TRUE, force = FALSE)
}