Documentation edits (#38)

* use `scales::label_percent()` instead of superseded `scales::percent_format()`.

* Update documentation and add more cross-references

* add backticks for better cross linking on website

* add roxygen md docs

---------

Co-authored-by: Samer Mouksassi <samermouksassi@gmail.com>

DESCRIPTION

@@ -69,3 +69,4 @@ LazyData: true
 VignetteBuilder: knitr
 RoxygenNote: 7.3.0
 Encoding: UTF-8
+Roxygen: list(markdown = TRUE)

NAMESPACE

@@ -34,12 +34,12 @@ importFrom(rlang,.data)
 importFrom(rlang,sym)
 importFrom(scales,identity_trans)
 importFrom(scales,label_parse)
+importFrom(scales,label_percent)
 importFrom(scales,label_wrap)
 importFrom(scales,math_format)
 importFrom(scales,muted)
 importFrom(scales,parse_format)
 importFrom(scales,percent)
-importFrom(scales,percent_format)
 importFrom(scales,trans_breaks)
 importFrom(scales,trans_format)
 importFrom(stats,median)

NEWS.md

@@ -1,22 +1,24 @@
 # ggquickeda 0.3.1
+
 * rebuilt docs fixed broken link
 * fixing some regressions introduced by bookmarking
 * fixing some errors by the new R rules for logical comparison length with `|` versus `||`
-* fixing error in aspect.ratio not numeric (was also fixed in ggplot2)
-* fixing other regressions with ggpairs color and group
+* fixing error in `aspect.ratio` not numeric (was also fixed in ggplot2)
+* fixing other regressions with `GGally::ggpairs()` color and group
 * added support of pasting values of two/three variables where it was just a placeholder
 * added function `ggkmrisktable()` to add facetable risk tables, km and by ntiles split exposures
-* added function `gglogisticexpdist()` and `ggcontinuousexpdist()` to add facetable logistic/linear regression with exposures split to  ntiles with exposure distributions by dose/group and optional y axis projections
+* added function `gglogisticexpdist()` and `ggcontinuousexpdist()` to add facetable logistic/linear regression with exposures split to ntiles with exposure distributions by dose/group and optional y axis projections
 * added link to github pages in description thanks to @olivroy
 * Fix part of ggplot2 3.4.0 deprecation warnings. (#27, @olivroy)
 
 # ggquickeda 0.3.0
+
 * added options to drop levels for user discrete shape and linetype scales
-* fixed ggpairs group input and added ability to combine y variables
+* fixed `GGally::ggpairs()` group input and added ability to combine y variables
 * fixed issue with barplot flip applied to bivariate plots
 * fixed issues with barplot percent labels when position is fill
 * added checkboxes to parse x and y axis labels text
-* added ability to regenerate state of apps to have a better UX workflow in Phoenix (James Craig)
+* added ability to regenerate state of apps to have a better UX workflow in Phoenix (James Craig, #28)
 * removed Welcome message and added padding
 * James Craig added as a co-author
 
@@ -28,7 +30,7 @@
 * restored target window annotations to work with factor/character x axis
 * added the possibility to change the size and linetype of density plots
 * added the possibility to choose the shape of boxplot outlier points
-* added options for x and y scale tick formatting are now visible when pairs plot are active
+* added options for `x` and `y` scale tick formatting are now visible when pairs plot are active
 * added options for identity line and for placement of custom/target text at plot edges
 * added checkbox to recode numeric variables with -99 to missing
 * removed ggstance dependency by adding needed positions to global.r
@@ -143,31 +145,32 @@
 * added possibility to divide multiple columns by a specified one (e.g. dose normalization)
 * "Treat as Categories:" now accepts character variables in addition to numeric.
 * "Treat as Numeric:" added to enable converting columns read as factors.
-* updated the docs for geom and stat_km stat_kmband stat_kmticks.
+* updated the docs for geom and `stat_km()` `stat_kmband()` `stat_kmticks()`.
 
 
 # ggquickeda 0.1.3
+
 * UI and options improvements for KM (line sizes, transparency, ignore group and color), linetype mappping
-* changed default error in stat_kmbands, borrowed stairstepn from ggalt and applying zoo::na.locf to deal with na
+* changed default error in `stat_kmbands()`, borrowed stairstepn from ggalt and applying `zoo::na.locf()` to deal with `NA`.
 * added the possibility to add median survival with or without CI
-* Using stat_cor from ggpubr to compute correlations for more flexibility and margin in facets
+* Using `ggpubr::stat_cor()` to compute correlations for more flexibility and margin in facets
 * added ability to ignore grouping and or ignore color mapping for correlation coefficient
 * added possibility to dodge means and medians
 * added possibility to specify means and medians N and Values transparency and a default seed for text_repel
-* geom text,label, withe repel variants and position dodge for median and mean labels
+* geom text,label, with repel variants and position dodge for median and mean labels
 * added transparency control for densities and histograms as well as linetype mappping
 * added binwidth control for histogram (user defined or auto), and position dodge
 * added window adjustment control for densities
 * linetypes for predefined quantiles
 * added the possibility to add one more therapeutic window and fixed a bug with categorical x axis
 * added possibility to remove x/y axis tick labels
 * fixes for continuous scales logic
-* Added initial support for ggpairs
+* Added initial support for `GGally::ggpairs()`
 
 
 
 # ggquickeda 0.1.2
-* bug fixed where in some cases plotdata did not apply filtering
+* bug fixed where in some cases plot data did not apply filtering
 * updated and expanded merge multiple levels of a factor
 * added mappings by shape for points and linetypes of lines and associated custom legends and possibility to ignore it
 * added the possibility of choosing more linetypes for lines
@@ -177,7 +180,7 @@
 * fixed bug in median line not honoring transparency when Median\PI selected with some UI tweaks
 * added more options for boxplot outliers (size and transparency)
 * separate color control for major and minor gridlines with the possibility to remove them
-* added annotation_logticks
+* added `annotation_logticks()`
 * make faceting less prone to fail when user mistakenly choose the same variable in rows and columns
 * make user defined color/fill scales
 

R/data.R

@@ -17,4 +17,7 @@
 #'   \item{AGECAT}{age category of the subject, a variable cutting Age into two values 0/1}
 #' }
 #' @source "sd_oral_richpk" from 'PKPDmisc' R package with an additional AGECAT variable
-"sample_data"
+#' @examples
+#' sample_data
+#' 
+"sample_data"

R/ggcontinuousexpdist.R

@@ -23,9 +23,10 @@ summary_df_cont <- function(x,y, probs = c(0.25,0.75,0.90,0.10)) {
 }
 plogis <- function(x) exp(x)/(1+exp(x))
 
-#' ggcontinuousexpdist
-#'
+#' Create a continuous exposure fit plot
+#' 
 #' Produces a logistic fit plot with a facettable exposures/quantiles/distributions in ggplot2
+#' 
 #' @param data Data to use with multiple endpoints stacked into Endpoint(endpoint name), response 0/1
 #' @param response name of the column holding the valuesresponse 0/1
 #' @param endpoint name of the column holding the name/key of the endpoint default to `Endpoint`
@@ -41,19 +42,19 @@ plogis <- function(x) exp(x)/(1+exp(x))
 #' @param ylab text to be used as y axis label
 #' @param mean_text_size mean text size default to 5
 #' @param mean_obs_bydose observed mean by dose TRUE/FALSE
-#' @param N_text_size N responders/Ntotal by exposure bin text size default to 5
+#' @param N_text_size N respondents/Ntotal by exposure bin text size default to 5
 #' @param binlimits_text_size 5 binlimits text size
 #' @param binlimits_ypos binlimits y position default to 0 
 #' @param binlimits_color binlimits text color default to "gray70"
 #' @param dist_position_scaler space occupied by the distribution default to 0.2 
 #' @param dist_offset offset where the distribution position starts 0
 #' @param lineranges_ypos where to put the lineranges -1
 #' @param lineranges_dodge lineranges vertical dodge value 1
-#' @param yproj project the probabilities on y axis TRUE/FALSE
+#' @param yproj project the probabilities on y axis `TRUE`/`FALSE`
 #' @param yproj_xpos y projection x position 0
 #' @param yproj_dodge  y projection dodge value 0.2
 #' @param yaxis_position where to put y axis "left" or "right"
-#' @param facet_formula facet formula to be use otherwise endpoint ~ expname
+#' @param facet_formula facet formula to be use otherwise `endpoint ~ expname`
 #' @param theme_certara apply certara colors and format for strips and default colour/fill
 #' @examples
 #' # Example 1

R/ggkm.R

@@ -11,7 +11,7 @@
 #' @importFrom scales trans_breaks
 #' @importFrom scales trans_format
 #' @importFrom scales math_format
-#' @importFrom scales percent_format
+#' @importFrom scales label_percent
 #' @importFrom scales parse_format
 #' @importFrom scales percent
 #' @importFrom scales label_parse
@@ -121,21 +121,22 @@ GeomKmticks <- ggplot2::ggproto("GeomKmticks", ggplot2::Geom,
 #' \code{geom_km} understands the following aesthetics (required aesthetics
 #' are in bold):
 #' \itemize{
-#'   \item \strong{\code{x}} The survival/censoring times. This is automatically mapped by \link{stat_km}
-#'   \item \strong{\code{y}} The survival probability estimates. This is automatically mapped by \link{stat_km}
-#'   smallest level in sort order is assumed to be 0, with a warning
+#'   \item \strong{\code{x}} The survival/censoring times. This is automatically mapped by [stat_km()]
+#'   \item \strong{\code{y}} The survival probability estimates. This is automatically mapped by [stat_km()]
+#'   smallest level in sort order is assumed to be 0, with a warning.
 #'   \item \code{alpha}
 #'   \item \code{color}
 #'   \item \code{linetype}
 #'   \item \code{size}
 #' }
 #'
 #' @inheritParams ggplot2::geom_point
-#' @seealso The default stat for this geom is \code{\link{stat_km}} see
+#' @seealso The default stat for this geom is [stat_km()] see
 #'   that documentation for more options to control the underlying statistical transformation.
 #' @export
 #' @examples
 #' library(ggplot2)
+#' set.seed(123)
 #' sex <- rbinom(250, 1, .5)
 #' df <- data.frame(time = exp(rnorm(250, mean = sex)), status = rbinom(250, 1, .75), sex = sex)
 #' ggplot(df, aes(time = time, status = status, color = factor(sex))) + geom_km()
@@ -157,8 +158,8 @@ geom_km <- function(mapping = NULL, data = NULL, stat = "km",
 #' \code{geom_kmband} understands the following aesthetics (required aesthetics
 #' are in bold):
 #' \itemize{
-#'   \item \strong{\code{x}} The survival/censoring times. This is automatically mapped by \link{stat_kmband}
-#'   \item \strong{\code{y}} The survival probability estimates. This is automatically mapped by \link{stat_kmband}
+#'   \item \strong{\code{x}} The survival/censoring times. This is automatically mapped by [stat_kmband()]
+#'   \item \strong{\code{y}} The survival probability estimates. This is automatically mapped by [stat_kmband()]
 #'   smallest level in sort order is assumed to be 0, with a warning
 #'   \item \code{alpha}
 #'   \item \code{color}
@@ -167,7 +168,7 @@ geom_km <- function(mapping = NULL, data = NULL, stat = "km",
 #' }
 #'
 #' @inheritParams ggplot2::geom_point
-#' @seealso The default stat for this geom is \code{\link{stat_kmband}} see
+#' @seealso The default stat for this geom is [stat_kmband()]. See
 #'   that documentation for more options to control the underlying statistical transformation.
 #' @export
 #' @examples
@@ -196,8 +197,8 @@ geom_kmband <- function(mapping = NULL, data = NULL, stat = "kmband",
 #' \code{geom_kmticks} understands the following aesthetics (required aesthetics
 #' are in bold):
 #' \itemize{
-#'   \item \strong{\code{x}} The survival/censoring times. This is automatically mapped by \link{stat_km}
-#'   \item \strong{\code{y}} The survival probability estimates. This is automatically mapped by \link{stat_km}
+#'   \item \strong{\code{x}} The survival/censoring times. This is automatically mapped by [stat_kmticks()]
+#'   \item \strong{\code{y}} The survival probability estimates. This is automatically mapped by [stat_kmticks()]
 #'   smallest level in sort order is assumed to be 0, with a warning
 #'   \item \code{alpha}
 #'   \item \code{color}
@@ -206,10 +207,9 @@ geom_kmband <- function(mapping = NULL, data = NULL, stat = "kmband",
 #' }
 #'
 #' @inheritParams ggplot2::geom_point
-#' @seealso The default stat for this geom is \code{\link{stat_kmticks}} see
+#' @seealso The default stat for this geom is [stat_kmticks] see
 #'   that documentation for more options to control the underlying statistical transformation.
 #' @export
-#' @rdname geom_kmticks
 #' @examples
 #' library(ggplot2)
 #' sex <- rbinom(250, 1, .5)
@@ -337,7 +337,7 @@ StatKmband <- ggplot2::ggproto("StatKmband", ggplot2::Stat,
 #'   to "identity". Other options include "event", "cumhaz", "cloglog", or
 #'   define your own using \link{trans_new}.
 #' @param firstx,firsty the starting point for the survival curves. By default,
-#'   the plot program obeys tradition by having the plot start at (0,1).
+#'   the plot program obeys tradition by having the plot start at `(0,1)`.
 #' @param start.time numeric value specifying a time to start calculating survival information.
 #'  The resulting curve is the survival conditional on surviving to start.time.
 #' @param type an older argument that combined stype and ctype, now deprecated. Legal values were
@@ -412,7 +412,7 @@ stat_km <- function(mapping = NULL, data = NULL, geom = "km",
 #' @inheritParams ggplot2::stat_identity
 #' @param trans Transformation to apply to the survival probabilities. Defaults
 #'   to "identity". Other options include "event", "cumhaz", "cloglog", or
-#'   define your own using \link{trans_new}.
+#'   define your own using [scales::trans_new()].
 #' @param firstx,firsty the starting point for the survival curves. By default,
 #'   the plot program obeys tradition by having the plot start at (0,1).
 #' @inheritParams survival::survfit.formula

R/ggkmrisktable.R

@@ -131,7 +131,7 @@ lung_long$facetdum <- "(all)"
 .trim <- function(x){gsub("^\\s+|\\s+$", "", x)}
 # from survminer
 
-#' ggkmrisktable
+#' Create a Kaplan-Meier plot with risk table 
 #'
 #' Produces a km plot with a facettable risk table in ggplot2
 #' 
@@ -183,10 +183,10 @@ lung_long$facetdum <- "(all)"
 #'  dplyr::filter(!is.na(ph.karno))|>
 #'  dplyr::filter(!is.na(pat.karno))|>
 #'  dplyr::filter(!is.na(ph.ecog))
-#'lung_long$ph.ecog <- ifelse(lung_long$ph.ecog>1,2,lung_long$ph.ecog)
-#'lung_long$ph.ecog <- as.factor(lung_long$ph.ecog )
-#'lung_long$ph.ecog <- as.factor(lung_long$ph.ecog )
-#'lung_long$facetdum <- "(all)"
+#' lung_long$ph.ecog <- ifelse(lung_long$ph.ecog>1,2,lung_long$ph.ecog)
+#' lung_long$ph.ecog <- as.factor(lung_long$ph.ecog )
+#' lung_long$ph.ecog <- as.factor(lung_long$ph.ecog )
+#' lung_long$facetdum <- "(all)"
 #' 
 #' ggkmrisktable(data = lung_long, time= "time", status ="DV",
 #'              exposure_metrics =c("age","ph.karno"),

R/gglogisticexpdist.R

@@ -23,11 +23,11 @@ summary_df <- function(x,y, probs = c(0.25,0.75,0.90,0.10)) {
 }
 plogis <- function(x) exp(x)/(1+exp(x))
 
-#' gglogisticexpdist
+#' Create a logistic fit plot
 #'
 #' Produces a logistic fit plot with a facettable exposures/quantiles/distributions in ggplot2
 #' @param data Data to use with multiple endpoints stacked into Endpoint(endpoint name), response 0/1
-#' @param response name of the column holding the valuesresponse 0/1
+#' @param response name of the column holding the values response 0/1
 #' @param endpoint name of the column holding the name/key of the endpoint default to `Endpoint`
 #' @param DOSE name of the column holding the DOSE values default to `DOSE`
 #' @param color_fill name of the column to be used for color/fill default to DOSE column
@@ -40,7 +40,7 @@ plogis <- function(x) exp(x)/(1+exp(x))
 #' @param xlab text to be used as x axis label
 #' @param ylab text to be used as y axis label
 #' @param prob_text_size probability text size default to 5
-#' @param prob_obs_bydose observed probability by dose TRUE/FALSE
+#' @param prob_obs_bydose observed probability by dose `TRUE`/`FALSE`
 #' @param N_text_size N responders/Ntotal by exposure bin text size default to 5
 #' @param binlimits_text_size 5 binlimits text size
 #' @param binlimits_ypos binlimits y position default to 0 
@@ -49,11 +49,11 @@ plogis <- function(x) exp(x)/(1+exp(x))
 #' @param dist_offset offset where the distribution position starts 0
 #' @param lineranges_ypos where to put the lineranges -1
 #' @param lineranges_dodge lineranges vertical dodge value 1
-#' @param yproj project the probabilities on y axis TRUE/FALSE
+#' @param yproj project the probabilities on y axis `TRUE`/`FALSE`
 #' @param yproj_xpos y projection x position 0
 #' @param yproj_dodge  y projection dodge value 0.2
 #' @param yaxis_position where to put y axis "left" or "right"
-#' @param facet_formula facet formula to be use otherwise endpoint ~ expname
+#' @param facet_formula facet formula to be use otherwise `endpoint ~ expname`
 #' @param theme_certara apply certara colors and format for strips and default colour/fill
 #' @examples
 #' # Example 1

R/logisticdata.R

@@ -21,4 +21,6 @@
 #'   \item{ICGI7}{response 1 to 7}
 #' }
 #' @source inspired from a real data submission
+#' @examples
+#' logistic_data
 "logistic_data"

inst/shinyapp/server.R

@@ -6325,7 +6325,7 @@ function(input, output, session) {
             p <- p +
               scale_y_continuous(expand = expansionobjy,
                                  breaks = waiver(),
-                                 labels = scales::percent_format())
+                                 labels = scales::label_percent())
           }
           if(input$yaxisformat=="scientificy"){
             p <- p +
@@ -6348,7 +6348,7 @@ function(input, output, session) {
             p <- p +
               scale_x_continuous(expand = expansionobjx,
                                  breaks = waiver(),
-                                 labels = scales::percent_format())
+                                 labels = scales::label_percent())
           }
           if(input$xaxisformat=="scientificx"){
             p <- p +
@@ -6449,7 +6449,7 @@ function(input, output, session) {
           input$yaxisformat=="percenty"){ 
         if (!input$addrisktable){
           p <- p  + 
-            scale_y_continuous(labels=scales::percent_format(), expand = expansionobjy)
+            scale_y_continuous(labels=scales::label_percent(), expand = expansionobjy)
         }#norisktable
         if (input$KM!="None" && input$addrisktable){
             p  <- p +
@@ -6526,7 +6526,7 @@ function(input, output, session) {
           !input$customytickslabel &&
           input$yaxisformat=="percenty" ) {
         p <- p  + 
-          scale_y_continuous(labels=scales::percent_format(),
+          scale_y_continuous(labels=scales::label_percent(),
                              breaks=as.numeric(unique(unlist (strsplit(input$yaxisbreaks, ","))) ),
                              minor_breaks = as.numeric(unique(unlist (strsplit(input$yaxisminorbreaks, ","))) ) ,
                              expand = expansionobjy) 
@@ -6637,12 +6637,12 @@ function(input, output, session) {
           input$xaxisformat=="percentx"){
         if(!input$customxticks){
           p <- p  + 
-            scale_x_continuous(labels=scales::percent_format() ,
+            scale_x_continuous(labels=scales::label_percent() ,
                                expand = expansionobjx) 
         }
         if(input$customxticks){
           p <- p  + 
-            scale_x_continuous(labels=scales::percent_format(),
+            scale_x_continuous(labels=scales::label_percent(),
                                breaks=as.numeric(unique(unlist (strsplit(input$xaxisbreaks, ","))) ),
                                minor_breaks = as.numeric(unique(unlist (strsplit(input$xaxisminorbreaks, ","))) ),
                                expand = expansionobjx)