predict.ProbDistr: Predict function for probability distributions in Methyl-IT
In genomaths/MethylIT.utils: MethylIT utility

View source: R/predictProbDistr.R View source: R/predictProbDistr.R

predict.ProbDistr

R Documentation

Predict function for probability distributions in Methyl-IT

Description

This is an utility function to get predictions from the probability distributions models used in Methyl-IT: Weibull, Gamma, and generalized Gamma. Some times, after the nonlinear fit of any of the mentioned modelsm we would like to evaluate the model output.

Usage

predict.ProbDistr(nlm, pred = "quant", q = 0.95, dist.name)

predict.ProbDistrList(
  nlm,
  pred = "quant",
  q = 0.95,
  dist.name,
  num.cores = 1L,
  tasks = 0L
)

Arguments

`nlm`	An object carrying the best nonlinear fit for a distribution model obtained with function `nonlinearFitDist`.
`pred`	Type of prediction resquested: density ('dens'),quantiles ('quant'), random number ('rnum') or probabilities ('prob').
`q`	numeric vector of quantiles, probabilities or an interger if pred = 'rnum'.
`dist.name`	name of the distribution to fit: Weibull2P (default: 'Weibull2P'), Weibull three-parameters (Weibull3P), gamma with three-parameter (Gamma3P), gamma with two-parameter (Gamma2P), generalized gamma with three-parameter ('GGamma3P') or four-parameter ('GGamma4P').
`num.cores, tasks`	Paramaters for parallele computation using package `BiocParallel-package`: the number of cores to use, i.e. at most how many child processes will be run simultaneously (see `bplapply` and the number of tasks per job (only for Linux OS).

Details

Predictions are based on the best model fit returned by function nonlinearFitDist. The possible prediction are: *density*, *quantiles*, *random number* or *probabilities*.

Examples

set.seed(1)
num.points <- 1000
HD <- makeGRangesFromDataFrame(
  data.frame(chr = 'chr1', start = 1:num.points, end = 1:num.points,
            strand = '*',
            hdiv = rweibull(1:num.points, shape = 0.75, scale = 1)),
  keep.extra.columns = TRUE)
nlms <- nonlinearFitDist(list(HD), column = 1, verbose = FALSE)

x=seq(0.1, 10, 0.05)
y <- predict(nlms[[1]], pred='dens', q = x,
                dist.name='Weibull2P')
y1 <- dweibull(x, shape = 0.75, scale = 1)
# The maximum difference between the 'theoretical' and estimated densities
max(abs(round(y, 2) - round(y1, 2)))

genomaths/MethylIT.utils documentation built on July 4, 2023, 12:05 a.m.