Add data simulation functions (#126)

* remove serodata .Rdata and .Rd files * remove R/serodata.R * doc: update functions documentation replacing for in examples * doc: update README and vignettes This commit changes the removed preloaded dataset `serodata` for the identical `chagas2012`. * fix: minor correction to test_modelling * feat: add function `get_foi_central_estimates()` to the modelling module. This change is meant to simplify `fit_seromodel()`. This commit also changes the name of the stanfit object in the output of `fit_seromodel()` from `fit` to `seromodel_fit`. * fix: minor corrections to tests according to commit 052c90c * fix: change to in model_comparison * doc: add documentation file for * doc: minor correction to documentation * refac: updates * refac: reduce the output of * fix(tests): change storing format of model object files from .json to .RDS and updates test_visualization accordingly * Remove `serodata` from pre-loaded datasets of the package (#107) * remove serodata .Rdata and .Rd files * remove R/serodata.R * doc: update functions documentation replacing for in examples * doc: update serofoi logo * doc: update README and vignettes This commit changes the removed preloaded dataset `serodata` for the identical `chagas2012`. * fix: minor correction to test_modelling * Fixes #115 (#116) * 82 tidyverse dependency (#120) * Enabling CMD check when doing pull request on `dev` * Removed tidyverse dependency * refac: `fit_seromodel` returns stanfit object (#117) * refac: `fit_seromodel` returns a stanfit object `fit_seromodel` returns a stanfit object now, which induces changes all across the pachage - `seromodel_object$seromodel_fit -> seromodel_object`: * `get_foi_central_estimates` * `extract_seromodel_summary` * `plot_seroprev_fitted` * `plot_foi` * `plot_rhats` * `plot_seromodel` * `get_table_rhats` - add `serodata` as input: * `get_foi_central_estimates` * `extract_seromodel_summary` * `plot_seroprev_fitted` * `plot_seromodel` * `plot_rhats` * `plot_foi` * `get_table_rhats` - use of `exposure_years` (this suggest adding a function for this purpose): * `get_foi_central_estimates` * `get_table_rhats` - minor corrections: * `plot_seromodel` -> correction to exception * tests(refac): re-generate .RDS files used in tests * tests(refac): update tests according to 7be19f9 - change `model_object$seromodel_fit` to `model_object` - add `serodata` as input where needed - minor syntax changes to test_issue_47.R * fix examples Fix examples of functions across the package in accordance to the new output of `fit_seromodel`. Updated examples: - `get_table_rhats` - `get_foi_central_estimates` - `get_prev_expanded` - `extract_seromodel_summary` - `plot_seroprev_fitted` - `plot_foi` - `plot_rhats` - `plot_seromodel` - `plot_info_table` * refac: replace `get_exposure_ages` for `get_cohort_ages` add `get_cohort_ages`: this function serves as replacement for `get_exposure_ages` and also includes the information of `exposure_years`, which was being used in some functions. Affected functions: - `fit_seromodel` - `get_exposure_matrix` - `get_table_rhats` * test: add test for `get_cohort_ages` to the test_modelling.R * refac: replace `serodata` for `cohort_ages` where needed Affected functions: - `get_table_rhats` - `get_foi_central_estimates` - `extract_seromodel_summary` - `plot_foi` - `plot_rhats` - `plot_seromodel` * doc: update documentation use @inheritParams to simplify documentation * add comment to exception in `fit_seromodel` * refac: specify argument `parameter_name` Explicitely specify argument `parameter_name = "logLikelihood"` in `loo::loo` * change condition to check whether sampling is done or not * doc: update vignettes according to changes * doc: update documentation and examples * doc: parameter description correction parameter `seromodel_object` description correction in `get_foi_central_estimates` * doc: minor corrections to functions documentation * fix: add input check for `age_mean_f` column (#123) add input check for `age_mean_f` in `serodata` in the following functions - `prepare_serodata` - `prepare_bin_data` * Add input validation (#92) * Added argument matching in fit_seromodel() * Added allowed options to arg foi_model * Added input validation to prepare_serodata() * Fixed the check for required columns * Regenerated the docs --------- Co-authored-by: ntorresd <[email protected]> * Remove test_visualisation and related files * Convert logo to png To keep a reasonable repo size * Add functions to simulate datasets to the seroprevalence_data module. (#58) * fix: add exception to function plot_foi() to plot a FOI trend with different length along with the data for the case when their sizes don't coincide * clean test_visualisation * feature: add three functions to simulate datasets. get_sim_counts() generates a list with simulated counts by age following a binomial distribution. generate_sim_data() uses the counts generated by get_sim_counts() to create a dataframe with the necessary structure to use other functions of the package. group_sim_data() serves to group the previously generated dataset by age group; right now it groups the data by periods of 5 years. * add test_simulate_data to test the data simulation functions in the seroprevalence_data module * doc: update generate_sim_data description * doc: complete get_sim_counts documentation and remove corresponding TODO item in seroprevalence_data.R * feat: add function get_sim_prob to seroprevalence_data.R. This function generates a list containing a seropositivity destribution by age following the specified force of infection. * doc: complete documentation of generate_sim_data in seroprevalence_data.R. Remove corresponding TODO item. * feat: add function `get_age_group` to seroprevalence_data.R. This function was taken from vaccineff [vaccineff](https://github.com/epiverse-trace/vaccineff). Minor changes to `group_sim_data` description. * feat (refac): use `get_age_group` function to generate the age groups in `group_sim_data` * doc: add or update documentation for data simulation functions * refac: minor changes to `get_sim_prob()` Move seed setting from `get_sim_prob()` to `get_sim_counts()`. Remove unnecessary parameters from documentation in `get_sim_prob()` * doc: update documentation for data simulation functions update `foi` parameter description - `get_sim_prob()` - `get_sim_counts` - `generate_sim_data` update `birth_year` parameter description - `prepare_serodata()` - `prepare_bin_data()` - `get_sim_counts()` - `get_sim_prob()` update function descriptions - `get_sim_counts()` - `generate_sim_data()` update return description - `get_sim_counts()` remove export - `prepare_bin_data()` - `get_sim_prob()` remove unnecessary input parameters - `get_sim_prob()` - `get_sim_counts()` * refac: change `size_age_class` for `sample_size_by_age` * doc: update documentation of data simulation functions update @return description for `get_age_group()` update documentation files for - `generate_sim_data()` - `get_age_group()` - `get_sim_counts()` - `group_sim_data()` * doc: remove example for `get_age_group()` * refac: change name and output of `get_sim_prob` change name of `get_sim_prob` to `get_sim_n_seropositive`. The function now returns a dataframe containing the exposure ages and the number of seropositive cases for each age. * doc: add example for `get_sim_counts()` * Revert "refac: change name and output of `get_sim_prob`" This reverts commit 966e463. * refac: change name and output of `get_sim_prob()` change name of `get_sim_prob()` to `get_sim_probability`. The output of the function is now a dataframe. * refac: change name and output of `get_sim_counts` change name of function `get_sim_counts` to `get_sim_n_seropositive` and update the documentation accordingly. minor correction to documentation of `get_sim_probability` * doc: minor correction to function documentation minor correction to function `get_age_group` documentation * doc: remove function `group_sim_data` from export remove function `group_sim_data` from export and the corresponding example * refac: simplify function `get_age_group` Remove unnecessary parameters from `get_age_group`. Now the function takes an age vector as input rather than a dataframe containing an specific age column. * test: remove data writing from test Remove data writing from `test_simulate_data.R` as well as the corresponding .csv files. * clean: remove unused data paths from test Remove unused data paths from `test_simulate_data` * test: add tests for `get_sim_` functionalities Add tests for `get_sim_probability` and `get_sim_n_seropositive` * test(refac): modify test for data sim functions Modify test for `generate_sim_data` and `group_sim_data` functions. Remove unnecessary model running. Remove `expect_doppleganger` tests. * add test for values from `get_sim_probability` Add `get_sim_probability` to export to enable testing. Test the values of the probabilities. * test: add statistical test for `generate_sim_data` The idea of the test is to make sure that the foi trend used to simulate the data is in the confidence interval of a suitable model. Add possibility to test for: - constant FoI - smooth-descendent FoI Minor clean up of `plot_foi`. * add pracma to dependencies * add parameter descriptions in `generate_sim_data` * doc: update documentation * test: add analytical test for constant foi Add analytical test for constant foi for large sample sizes. * test: add test for data simulation with time-varying foi * small syntax corrections in test_sim_data.R --------- Co-authored-by: ntorresd <[email protected]> Co-authored-by: Nicolás Torres Domínguez <[email protected]> Co-authored-by: Jaime Pavlich-Mariscal <[email protected]> Co-authored-by: James Azam <[email protected]>
epiverse-trace · Nov 8, 2023 · 6db956b · 6db956b
1 parent fe428e5
commit 6db956b
Show file tree

Hide file tree

Showing 16 changed files with 509 additions and 34 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -56,6 +56,7 @@ Imports:
     jsonlite,
     loo,
     purrr,
+    pracma,
     qtl,
     reshape2,
     usethis,

diff --git a/NAMESPACE b/NAMESPACE
@@ -2,10 +2,13 @@
 
 export(extract_seromodel_summary)
 export(fit_seromodel)
+export(generate_sim_data)
 export(get_cohort_ages)
 export(get_exposure_matrix)
 export(get_foi_central_estimates)
 export(get_prev_expanded)
+export(get_sim_n_seropositive)
+export(get_sim_probability)
 export(get_table_rhats)
 export(plot_foi)
 export(plot_info_table)

diff --git a/R/seroprevalence_data.R b/R/seroprevalence_data.R
@@ -22,7 +22,9 @@
 #' \tabular{ll}{
 #' \code{age_mean_f} \tab Floor value of the average between age_min and age_max \cr \tab \cr
 #' \code{sample_size} \tab The size of the sample \cr \tab \cr
-#' \code{birth_year} \tab The year in which the individuals of each age group were bornt \cr \tab \cr
+#' \code{birth_year} \tab Years in which the subjects were borned
+#' according to the age group marker \code{age_mean_f}\cr \tab \cr
+#' according to the age group marker \code{age_mean_f}\cr \tab \cr
 #' \code{prev_obs} \tab Observed prevalence \cr \tab \cr
 #' \code{prev_obs_lower} \tab Lower limit of the confidence interval for the observed prevalence \cr \tab \cr
 #' \code{prev_obs_upper} \tab Upper limit of the confidence interval for the observed prevalence \cr \tab \cr
@@ -37,21 +39,24 @@ prepare_serodata <- function(serodata = serodata,
   #Check that serodata has the right columns
   stopifnot("serodata must contain the right columns" =
             all(c("survey", "total", "counts", "age_min", "age_max", "tsur",
-                  "country","test","antibody"
+                  "country", "test", "antibody"
                   ) %in%
                   colnames(serodata)
+                ) |
+            all(c("survey", "total", "counts", "age_mean_f", "tsur",
+                  "country", "test", "antibody") %in%
+                  colnames(serodata)
                 )
-            )
+  )
   if(!any(colnames(serodata) == "age_mean_f")){
     serodata <- serodata %>%
       dplyr::mutate(age_mean_f = floor((age_min + age_max) / 2), sample_size = sum(total))
-  }  
+  }
 
   if(!any(colnames(serodata) == "birth_year")){
     serodata <- serodata %>%
       dplyr::mutate(birth_year = .data$tsur - .data$age_mean_f)
   }
-
   serodata <- serodata %>%
     cbind(
       Hmisc::binconf(
@@ -76,8 +81,8 @@ prepare_serodata <- function(serodata = serodata,
 #' Function that prepares a pre-processed serological survey dataset to plot the binomial confidence intervals of the seroprevalence grouped by
 #' age group
 #'
-#' This function prepapares a given pre-processed serological dataset (see \code{\link{prepare_serodata}}) to plot the binomial confidence intervals 
-#' of its corresponding seroprevalence grouped by age group. 
+#' This function prepapares a given pre-processed serological dataset (see \code{\link{prepare_serodata}}) to plot the binomial confidence intervals
+#' of its corresponding seroprevalence grouped by age group.
 #' @inheritParams run_seromodel
 #' @return data set with the binomial confidence intervals
 #' @examples
@@ -115,4 +120,190 @@ prepare_bin_data <- function(serodata) {
     p_obs_bin_u = .data$Upper
   )
   return(xx)
-}
+}
+
+#' Function that generates the probabilities of being previously exposed to a pathogen.
+#'
+#' @param sim_data A dataframe object containing the following columns:
+#' \tabular{ll}{
+#' \code{tsur} \tab Year of the survey\cr \tab \cr
+#' \code{age_mean_f} \tab Age \cr \tab \cr
+#' \code{birth_year} \tab Years in which the subjects were borned
+#' according to the age group marker \code{age_mean_f}\cr \tab \cr
+#' }
+#' @param foi Numeric atomic vector corresponding to the desired Force-of-Infection ordered from past to present
+#' @param seed The seed for random number generation.
+#' @return A dataframe containing the following columns:
+#' \tabular{ll}{
+#' \code{age} \tab Exposure ages \cr \tab \cr
+#' \code{probability} \tab Probability to obtain a seropositive case for each age according to the provided FoI\cr \tab \cr
+#' }
+#' @export
+get_sim_probability <- function(sim_data, foi) {
+  cohort_ages <- get_cohort_ages(sim_data)
+  exposure_ages <- rev(cohort_ages$age)
+  exposure_matrix <- get_exposure_matrix(sim_data)
+  probabilities <- purrr::map_dbl(exposure_ages, ~1-exp(-pracma::dot(exposure_matrix[., ], foi)))
+
+  sim_probability <- data.frame(
+    age = exposure_ages,
+    probability = probabilities
+  )
+  return(sim_probability)
+}
+
+#' Function that generates a sample of counts of seropositive individuals by sampling from a binomial distribution
+#'
+#' @param sim_data A dataframe object containing the following columns:
+#' \tabular{ll}{
+#' \code{tsur} \tab Year of the survey\cr \tab \cr
+#' \code{age_mean_f} \tab Age \cr \tab \cr
+#' \code{birth_year} \tab Years in which the subjects were borned
+#' according to the age group marker \code{age_mean_f}\cr \tab \cr
+#' }
+#' @param foi Numeric atomic vector corresponding to the desired Force-of-Infection ordered from past to present
+#' @param sample_size_by_age Sample size for each age group: either a single integer indicating that the sample size is 
+#' the same for all ages or a vector of sample sizes the same length as
+#' This corresponds to the number of trials \code{size} in \link[stats]{rbinom}.
+#' @param seed The seed for random number generation.
+#' @return A dataframe containing the following columns:
+#' \tabular{ll}{
+#' \code{age} \tab Age by the time of the survey \cr \tab \cr
+#' \code{n_seropositive} \tab Number of positive cases sampled according to the provided FoI \cr \tab \cr
+#' }
+#' simulated list of counts following a binomial distribution in accordance with a given
+#' force of infection and age class sizes.
+#' @examples
+#'\dontrun{
+#' foi <- rep(0.02, 50)
+#' sim_data <- generate_sim_data(foi = foi,
+#'                               sample_size_by_age = 5,
+#'                               tsur = 2050,
+#'                               birth_year_min = 2000,
+#'                               survey_label = 'foi_sim')
+#' sim_n_seropositive <- get_sim_n_seropositive(sim_data = sim_data,
+#'                                              foi = foi)
+#' }
+#' @export
+get_sim_n_seropositive <- function(sim_data, foi, sample_size_by_age, seed = 1234) {
+  sim_probability <- get_sim_probability(sim_data = sim_data, foi = foi)
+
+  set.seed(seed = seed)
+  n_seropositive <- purrr::map_int(sim_probability$probability, ~rbinom(1, sample_size_by_age, .))
+
+  sim_n_seropositive <- data.frame(
+    age = sim_probability$age,
+    n_seropositive = n_seropositive
+  )
+  return(sim_n_seropositive)
+}
+
+#' Function that generates a simulated serosurvey according to the specified FoI
+#'
+#' @param foi Numeric atomic vector corresponding to the desired Force-of-Infection ordered from past to present
+#' @param sample_size_by_age Size of each age group specified by either an atomic
+#' vector of the same size as \code{foi} or an integer.
+#' This corresponds to the number of trials \code{size} in \link[stats]{rbinom}.
+#' @param tsur Year in which the serosurvey was conducted.
+#' @param birth_year_min Minimum age of year in the simulated serosurvey.
+#' @param survey_label Label for the resulting simulated serosurvey.
+#' @return Dataframe object containing the simulated data generated from \code{foi}
+#' @examples
+#'\dontrun{
+#' sample_size_by_age = 5
+#' foi <- rep(0.02, 50)
+#' sim_data <- generate_sim_data(foi = foi,
+#'                               sample_size_by_age = sample_size_by_age,
+#'                               tsur = 2050,
+#'                               birth_year_min = 2000,
+#'                               survey_label = 'sim_constant_foi')
+#' }
+#' @export
+generate_sim_data <- function(foi,
+                              sample_size_by_age,
+                              tsur,
+                              birth_year_min,
+                              survey_label,
+                              test = "fake",
+                              antibody = "IgG",
+                              seed = 1234
+                              ){
+    sim_data <- data.frame(birth_year = c(birth_year_min:(tsur - 1))) %>%
+        mutate(tsur = tsur,
+            country = 'None',
+            test = test,
+            antibody = antibody,
+            survey = survey_label,
+            age_mean_f = tsur - birth_year)
+    sim_n_seropositive <- get_sim_n_seropositive(sim_data, foi, sample_size_by_age, seed = seed)
+    sim_data <- sim_data %>%
+        mutate(counts = sim_n_seropositive$n_seropositive,
+               total = sample_size_by_age) %>%
+        prepare_serodata()
+
+    return(sim_data)
+}
+
+#' Method for constructing age-group variable from age column
+#'
+#' This function was taken from \link[vaccineff]{get_age_group}.
+#' This method splits an age interval from age_min to age_max into
+#' (age_max-age_min)/step intervals.
+#' By default age_min is set 0, however it can be assigned by
+#' convenience.
+#' If the method finds ages greater or equal than age_max
+#' it assigns the string ">{age_max}".
+#' To avoid errors it is necessary to set step < age_max.
+#' It is also suggested to choose the step such
+#' that age_max%%(step+1) = 0.
+#' @param age vector containing age information
+#' @param  step step used to split the age interval
+#' @return age_group factor variable grouping \code{age} by the age intervals 
+#' specified by \code{min(age)}, \code{max(age)}.
+get_age_group <- function(age, step) {
+  age_min <- min(age)
+  age_max <- max(age)
+  n_steps <- as.integer((age_max - age_min) / step) + 1
+  limits_low <- c(as.integer(seq(age_min,
+                                 age_max,
+                                 length.out = n_steps)))
+  limits_hgh <- limits_low + step
+  lim_labels <- paste(as.character(limits_low), as.character(limits_hgh),
+                      sep = "-")
+  lim_labels[length(lim_labels)] <- paste0("+",
+                                           limits_low[length(limits_low)])
+  lim_breaks <- c(-Inf, limits_low[2:length(limits_low)] - 1, Inf)
+
+  age_group <- cut(age,
+                   breaks = lim_breaks,
+                   labels = lim_labels,
+                   # this is for the intervals to be closed to the left and open to the right
+                   right = FALSE)
+  return(age_group)
+}
+
+#' Function that groups a simulated serological dataset by age
+#'
+#' @param sim_data Dataframe with the same structure as the output of \code{\link{generate_sim_data}}.
+#' @param col_age name of the column containing the age information
+#' @param step step used to split the age interval
+#' @return Dataframe object containing grouped simulated data generated from \code{foi}
+group_sim_data <- function(sim_data,
+                          col_age = "age_mean_f",
+                          step = 5) {
+    age <- sim_data[[col_age]]
+    sim_data$age_group <-  get_age_group(age = age, step = step)
+    sim_data_grouped <- sim_data %>%
+      group_by(age_group) %>%
+    dplyr::summarise(total = sum(total), counts = sum(counts)) %>%
+      mutate(tsur = sim_data$tsur[1],
+              country = "None",
+              survey = sim_data$survey[1],
+              test = sim_data$test[1],
+              antibody = sim_data$antibody[1]) %>%
+      mutate(age_min = as.integer(sub("\\-.*", "", age_group)),
+              age_max = as.integer(sub(".*\\-", "", age_group))) %>%
+      prepare_serodata()
+
+    return(sim_data_grouped)
+}
diff --git a/R/visualisation.R b/R/visualisation.R
@@ -27,7 +27,7 @@ plot_seroprev <- function(serodata,
 #' Function that generates a seropositivity plot corresponding to the specified fitted serological model
 #'
 #' This function generates a seropositivity plot of the specified serological model object. This includes the original data grouped by age
-#' as well as the obtained fitting from the model implementation. Age is located on the x axis and seropositivity on the y axis with its 
+#' as well as the obtained fitting from the model implementation. Age is located on the x axis and seropositivity on the y axis with its
 #' corresponding confidence interval.
 #' @inheritParams get_foi_central_estimates
 #' @inheritParams run_seromodel
@@ -108,7 +108,7 @@ plot_seroprev_fitted <- function(seromodel_object,
 #' Function that generates a Force-of-Infection plot corresponding to the specified fitted serological model
 #'
 #' This function generates a Force-of-Infection plot from the results obtained by fitting a serological model.
-#' This includes the corresponding binomial confidence interval. 
+#' This includes the corresponding binomial confidence interval.
 #' The x axis corresponds to the decades covered by the survey the y axis to the Force-of-Infection.
 #' @inheritParams get_foi_central_estimates
 #' @param size_text Text size use in the theme of the graph returned by the function.
@@ -138,7 +138,6 @@ plot_foi <- function(seromodel_object,
       foi <- rstan::extract(seromodel_object,
                             "foi",
                             inc_warmup = FALSE)[[1]]
-
       #-------- This bit is to get the actual length of the foi data
       foi_data <- get_foi_central_estimates(seromodel_object = seromodel_object,
                                             cohort_ages = cohort_ages)
@@ -168,16 +167,23 @@ plot_foi <- function(seromodel_object,
         ggplot2::xlab("Year")
       #TODO Add warning for foi_sim of different length than exposure years
       if (!is.null(foi_sim)){
-        foi_data_length <- nrow(foi_data)
-        foi_sim_length <- length(foi_sim)
-        remove_x_values <- foi_sim_length - foi_data_length
-
-        foi_sim_data <- data.frame(year = foi_data$year, 
-                                  foi_sim = foi_sim[-c(1:remove_x_values)])    
-        foi_plot <- foi_plot + 
-          ggplot2::geom_line(data = foi_sim_data, ggplot2::aes(x = year, y = foi_sim),
-                            colour = "#b30909",
-                            size = size_text/8)
+        if (nrow(foi_data) != length(foi_sim)) {
+          remove_x_values <- length(foi_sim) - nrow(foi_data)
+          foi_sim_data <- data.frame(year = foi_data$year,
+                                    foi_sim = foi_sim[-c(1:remove_x_values)])
+          foi_plot <- foi_plot +
+            ggplot2::geom_line(data = foi_sim_data, ggplot2::aes(x = year, y = foi_sim),
+                              colour = "#b30909",
+                              size = size_text / 8)
+        }
+        else{
+          foi_sim_data <- data.frame(year = foi_data$year,
+                                    foi_sim = foi_sim)
+          foi_plot <- foi_plot +
+            ggplot2::geom_line(data = foi_sim_data, ggplot2::aes(x = year, y = foi_sim),
+                              colour = "#b30909",
+                              size = size_text / 8)
+        }
       }
     }
   } else {
@@ -208,8 +214,8 @@ plot_foi <- function(seromodel_object,
 
 #' Function that generates a plot of the R-hat estimates of the specified fitted serological model
 #'
-#' This function generates a plot of the R-hat estimates obtained for a specified fitted serological model \code{seromodel_object}. 
-#' The x axis corresponds to the decades covered by the survey and the y axis to the value of the rhats. 
+#' This function generates a plot of the R-hat estimates obtained for a specified fitted serological model \code{seromodel_object}.
+#' The x axis corresponds to the decades covered by the survey and the y axis to the value of the rhats.
 #' All rhats must be smaller than 1 to ensure convergence (for further details check \link[bayesplot]{rhat}).
 #' @inheritParams get_foi_central_estimates
 #' @param size_text Text size use in the theme of the graph returned by the function.
@@ -322,7 +328,7 @@ plot_seromodel <- function(seromodel_object,
       model_summary <- extract_seromodel_summary(seromodel_object = seromodel_object,
                                                  serodata = serodata)
       summary_table <- t(
-        dplyr::select(model_summary, 
+        dplyr::select(model_summary,
         c('foi_model', 'dataset', 'elpd', 'se', 'converged')))
       summary_plot <-
         plot_info_table(summary_table, size_text = size_text)
@@ -333,7 +339,7 @@ plot_seromodel <- function(seromodel_object,
         foi_plot,
         rhats_plot,
         ncol = 1,
-        nrow = 4, 
+        nrow = 4,
         rel_heights = c(0.5, 1, 1, 1)
       )
     }
@@ -358,7 +364,7 @@ plot_seromodel <- function(seromodel_object,
       ggplot2::ylab(" ") +
       ggplot2::xlab(" ")
     g1 <- g0
-    # TODO: This 
+    # TODO: This
     g0 <- g0 + ggplot2::labs(subtitle = seromodel_object$model_name) +
       ggplot2::theme(plot.title = ggplot2::element_text(size = 10))
 
@@ -369,7 +375,7 @@ plot_seromodel <- function(seromodel_object,
   return(plot_arrange)
 }
 
-# TODO Improve documentation of @return. 
+# TODO Improve documentation of @return.
 # TODO Give more details about the generated plot
 #' Function that generates a plot for a given table
 #'
@@ -399,4 +405,4 @@ plot_info_table <- function(info, size_text) {
                        fontface = "bold")
 
   return(p)
-}
+}
-Original file line number
+Diff line change
@@ Expand Up / @@ -56,6 +56,7 @@ Imports: @@
         jsonlite,
         loo,
         purrr,
+        pracma,
         qtl,
         reshape2,
         usethis,
@@ Expand Down @@