peppwR_up

DESCRIPTION

@@ -1,14 +1,14 @@
 Package: peppwR
 Title: Power Analysis For Phosphopeptide Abundance Hypothesis Test
-Version: 0.0.0.9000
+Version: 0.0.0.9001
 Authors@R: 
     person("Dan", "MacLean", , "dan.maclean@tsl.ac.uk", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-1032-0887"))
 Description: Estimate best fit distributions and do power analysis for hypothesis tests on phosphopeptide abundance data.
 License: MIT + file LICENSE
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.2.3
+RoxygenNote: 7.3.1
 URL: https://github.com/danmaclean/peppwR
 BugReports: https://github.com/danmaclean/peppwR/issues
 Imports: 

NAMESPACE

@@ -1,2 +1,15 @@
 # Generated by roxygen2: do not edit by hand
 
+export(avail_dists)
+export(d2n)
+export(do_fits)
+export(evaldist)
+export(fake_leg)
+export(fill_scale)
+export(find_fits)
+export(parse_fitdist)
+export(parse_univariateML)
+export(plot_best)
+export(plot_failed)
+export(single_fit)
+export(squash_fits)

R/fits.R

@@ -1,4 +1,14 @@
-
+#' Fit a Single Distribution
+#'
+#' @description This fits a specified distribution to the first column of a data frame.
+#'
+#' @param df The data frame containing the data to fit.
+#' @param dist The distribution name or a list of distribution models to try fitting. One of "nbinom", "gamma", "snorm", "norm",
+#'        "invgamma", "invgauss", "lnorm", "lgamma",  and "pareto"
+#'
+#'@return A list containing the fit result or the distribution name in case there is an error.
+#'
+#' @export
 single_fit <- function(df, dist){
   if (dist %in% c("nbinom")){
     result <- tryCatch(
@@ -16,7 +26,15 @@ single_fit <- function(df, dist){
 
 }
 
-
+#' Available Distributions
+#'
+#' @description This provides a vector of available distributions for fitting.
+#'
+#' @return A character vector of available distribution names.
+#'
+#'
+#'
+#' @export
 avail_dists <- function() {
   c("gamma", "norm",
     "snorm", "invgamma",
@@ -25,13 +43,35 @@ avail_dists <- function() {
     "nbinom")
 }
 
+
+#' Distribution Name to Normalized Name
+#'
+#' @description This converts distribution tags to names that humans can read and understand.
+#'
+#' @param tag A character string representing the distribution tag.
+#'
+#' @return A character string with the human-readable distribution name.
+#'
+#'
+#' @export
 d2n <- function(tag){
   v <- c("Gamma", "Normal", "Skew Normal", "InvGamma", "Inverse Gaussian",
            "Lognormal", "Log Gamma", "Pareto", "Negative Binomial")
   names(v) <- avail_dists()
   v[tag]
 }
 
+
+#' Fit Multiple Distributions
+#'
+#' @description This is used to fits multiple distributions to the first column of a data frame.
+#'
+#' @param df A data frame with the data to fit.
+#'
+#' @return A data frame with the fit results for each distribution.
+#'
+#'
+#' @export
 do_fits <- function(df) {
   dists <- avail_dists()
 
@@ -41,6 +81,16 @@ do_fits <- function(df) {
 
 }
 
+
+#' Squash Fit Results
+#'
+#' @description This processes fit results into a tidy format.
+#'
+#' @param fit A list containing the fit result.
+#'
+#' @return A tibble with the distribution name, log-likelihood, and Akaike Information Criterion (AIC).
+#'
+#' @export
 squash_fits <- function(fit){
 
   if (length(fit) == 1){
@@ -58,6 +108,18 @@ squash_fits <- function(fit){
 
 }
 
+
+
+#' Parse Fit Results from fitdistrplus
+#'
+#' @description This converts fit results from the `fitdistrplus` package into a tidy format.
+#'
+#' @param fit An object returned by `fitdistrplus::fitdist`.
+#'
+#' @return A tibble with the distribution name, log-likelihood, and AIC.
+#'
+#'
+#' @export
 parse_fitdist <- function(fit){
   tibble::tibble(
     dist = fit$distname,
@@ -66,6 +128,16 @@ parse_fitdist <- function(fit){
   )
 }
 
+
+#' Parse Fit Results from univariateML
+#'
+#' @description This converts fit results from the `univariateML` package into a tidy format.
+#'
+#' @param fit An object returned by `univariateML::model_select`.
+#'
+#' @return A tibble with the distribution name, log-likelihood, and AIC.
+#'
+#' @export
 parse_univariateML <- function(fit){
   tibble::tibble(
     dist = attr(fit, "model"),

R/peppwr.R

@@ -1,4 +1,15 @@
-
+#' Find Fits for Nested Data
+#'
+#' @description This groups the data frame by specified columns, nests the data, and applies the fitting functions to each group.
+#'
+#' @param df A data frame containing the data to be fitted.
+#' @param id_col A character string specifying the column name for the identifier. Default is "id".
+#' @param group_col A character string specifying the column name for the group. Default is "group".
+#' @param value_col A character string specifying the column name for the values to be fitted. Default is "value".
+#'
+#' @return A data frame with nested data and the fit results for each group.
+#'
+#' @export
 find_fits <- function(df, id_col="id", group_col="group", value_col="value"){
     tidyr::nest(df, .by = tidyr::all_of(c(id_col, group_col )), data = {{value_col}} ) |>
     dplyr::mutate(

R/plots.R

@@ -1,10 +1,32 @@
+
+#' Fill Scale for Distributions
+#'
+#' @description This generates a fill scale for distributions using the Set3 palette from RColorBrewer.
+#'
+#' @param name A character string specifying the name of the fill scale.
+#'
+#' @return A ggplot2 scale fill manual object.
+#'
+#'
+#' @export
 fill_scale <- function(name = "name"){
   cols <- RColorBrewer::brewer.pal(length(avail_dists()), "Set3")
   names(cols) <- d2n(avail_dists())
   ggplot2::scale_fill_manual(name = name, values = cols)
 }
 
 
+
+#' Plot Best Fitted Models
+#'
+#' @description This plots the best-fitted models based on Log-Likelihood and AIC from the fits data frame.
+#'
+#' @param fits_df A data frame containing the fit results.
+#' @param fit_col A character string specifying the column name that contains the nested fit results. Default is "fits".
+#'
+#' @return A ggplot object showing the best-fitted models.
+#'
+#' @export
 plot_best <- function(fits_df, fit_col="fits") {
   #TODO check and test df has fit_col 'fits'
 
@@ -49,6 +71,17 @@ plot_best <- function(fits_df, fit_col="fits") {
 }
 
 
+
+#' Plot Failed Models
+#'
+#' @description This plots the count of failed model fits from the fits data frame.
+#'
+#' @param fits_df A data frame containing the fit results.
+#' @param fit_col A character string specifying the column name that contains the nested fit results. Default is "fits".
+#'
+#' @return A ggplot object showing the count of failed models.
+#'
+#' @export
 plot_failed <- function(fits_df, fit_col = "fits"){
 
   if (! fit_col %in% colnames(fits_df)){
@@ -79,6 +112,17 @@ plot_failed <- function(fits_df, fit_col = "fits"){
 
 }
 
+
+
+#' Fake Legend
+#'
+#' @description This generates a fake legend for the distribution plots.
+#'
+#' @param name A character string specifying the name of the legend. Default is "Dist".
+#'
+#' @return A ggplot2 legend object.
+#'
+#' @export
 fake_leg <- function(name="Dist") {
   p <- data.frame(
     dist = d2n(avail_dists()),
@@ -93,6 +137,20 @@ fake_leg <- function(name="Dist") {
     cowplot::get_legend(p)
 }
 
+
+
+
+#' Evaluate Distributions
+#'
+#' @description This evaluates and plots the best-fitted and failed models from the fits data frame.
+#'
+#' @param fits_df A data frame containing the fit results.
+#'
+#' @param fit_col A character string specifying the column name that contains the nested fit results. Default is "fits".
+#'
+#' @return A ggplot object showing the evaluation of fitted models.
+#'
+#' @export
 evaldist <- function(fits_df, fit_col="fits"){
   a <- plot_best(fits_df, fit_col = fit_col)
   b <- plot_failed(fits_df, fit_col=fit_col)