inst/popkinr/nonmem_toolbox/ui.R
In pnolain/popkinr: Shiny applications for pharmacometrics
suppressPackageStartupMessages({
library(shiny)
library(shinydashboard)
library(DT)
library(shinyjs)
library(shinyAce)
library(shinytoastr)
library(rhandsontable)
library(ggplot2)
library(tidyr)
library(dplyr)
library(stringr)
library(purrr)
library(readr)
library(lubridate)
library(hms)
library(broom)
library(future)
library(pmxploit)
library(rlang)
library(xml2)
library(vpc)
library(shinycssloaders)
library(shinyTree)
library(jsonlite)
})
source("modules/extendedPlot.R")
source("modules/extendedDT.R")
source("modules/extendedDT_with_code.R")
# Show VPC tab only if a NONMEM executable path is available
vpc_tab <- (if(Sys.getenv("NM_EXE") != "") {
menuItem("VPC", tabName = "vpc", icon = icon("area-chart"))
} else {
NULL
})
app_title <- "PopkinR - NONMEM Toolbox"
smoothing_methods <- c(
"None" = "none",
"Linear model" = "lm",
"Local Polynomial Regression" = "loess"
)
# Edit header bar to integrate run path
db_header <-
dashboardHeader(title = app_title, titleWidth = 400)
db_header$children[[3]]$children[[3]] <-
uiOutput("application_top_text")
dashboardPage(
skin = "green",
title = app_title,
db_header,
dashboardSidebar(
sidebarMenu(
id = "main_menu",
menuItem(
"Home",
tabName = "home",
icon = icon("home"),
selected = TRUE
),
menuItem("Metadata", tabName = "metadata", icon = icon("info")),
menuItem("Population", tabName = "population", icon = icon("globe")),
menuItem("Diagnostics", tabName = "diagnostics", icon = icon("search")),
menuItem(
"Individuals",
icon = icon("users"),
menuSubItem("Parameters", tabName = "parameters_tab", icon = icon("cogs")),
menuSubItem(
"Covariates",
tabName = "covariates_tab",
icon = icon("pie-chart")
),
menuSubItem(
"Parameters vs Covariates",
tabName = "params_vs_covs_tab",
icon = icon("table")
)
),
menuItem(
"Quality criteria",
tabName = "quality_criteria",
icon = icon("check")
),
menuItem(
"Outliers detection",
tabName = "outliers",
icon = icon("user-secret")
),
vpc_tab,
menuItem(
"Compare to...",
tabName = "comparison",
icon = icon("sort-numeric-asc")
),
menuItem(
"Files",
icon = icon("copy"),
menuSubItem(
"Control stream",
tabName = "control_stream_file",
icon = icon("file-code")
),
menuSubItem("Report", tabName = "report_file", icon = icon("file-text")),
menuSubItem("Tables", tabName = "table_files", icon = icon("database"))
),
br(),
div(align = "center", actionButton("show_filters", "Filter data")),
hr(),
div(
align = "center",
img(
src = "media/pmxploit.svg",
width = 175,
height = 175
),
h6(
sprintf(
"PopkinR version: %s (%s)",
packageDescription("popkinr")$Version,
packageDescription("popkinr")$Date
)
),
h6("Patrick Nolain"),
tags$a(icon("github"), "github.com/pnolain", href="https://github.com/pnolain"),
hr(),
actionLink("example", "Demo")
)
)
),
dashboardBody(
useShinyjs(),
includeScript("www/js/sweetalert2.min.js"),
shiny::includeCSS("www/css/sweetalert2.min.css"),
includeScript("www/js/script.js"),
shiny::includeCSS("www/css/styles.css"),
useToastr(),
div(
id = "loading-content",
img(
src = "media/pmxploit_animated.svg",
width = 200,
height = 200
)
),
tabItems(
tabItem(tabName = "home",
fluidRow(
box(
id = "browsing_box",
width = 12,
title = "Select a run",
strong(actionLink("select_run", "Browse..."))
),
box(
id = "recent_box",
title = "History",
collapsible = TRUE,
width = 12,
uiOutput("open_last_run")
),
shinyjs::hidden(
div(id = "summary_box",
box(
title = "Summary",
width = 12,
h4("Estimation"),
fluidRow(
column(12,
DT::dataTableOutput("estimations_table"))
),
h4("Run information"),
fluidRow(
column(7,
uiOutput("run_information")),
column(
5,
fluidRow(
valueBoxOutput(width = 6, "subjects_box"),
valueBoxOutput(width = 6, "observations_box")
),
fluidRow(
valueBoxOutput(width = 6, "duration_box"),
valueBoxOutput(width = 6, "cpu_box")
),
fluidRow(valueBoxOutput(width = 6, "nm_version_box"))
)
)
))
)
)),
tabItem(
tabName = "metadata",
title = "Metadata",
fluidRow(
box(
width = 12,
title = "Metadata",
fluidRow(column(6,
tags$div(
tags$p(
"Metadata refer to details of the model allowing the user to manipulate several elements for plotting and reporting (e.g. compartments/parameters/covariates names, covariates types) by editing the following boxes."
),
tags$p("One can also:"),
tags$ul(
tags$li(
actionButton("browse_metadata", label = "Import"),
"metadata from another run"
),
tags$li(
actionButton("clear_metadata", label = "Clear"),
"(remove) metadata"
),
tags$li(
actionButton("reset_metadata", label = "Reset"),
"recent changes in metadata to their state at loading"
),
tags$li(
actionButton("save_metadata", label = "Save"),
"metadata to the run location"
)
)
)),
column(6, p()))
),
box(
width = 6,
title = "Compartments",
rHandsontableOutput("compartments_metadata")
),
box(
width = 6,
title = "Independent variables",
rHandsontableOutput("idv_metadata")
),
box(
width = 12,
title = "Parameters",
fluidRow(
width = 12,
column(
width = 4,
h4("Fixed effects (THETAs)"),
rHandsontableOutput("thetas_metadata")
),
column(
width = 4,
h4("Random effects (ETAs)"),
rHandsontableOutput("etas_metadata")
),
column(
width = 4,
h4("Individual parameters (POSTHOCs)"),
rHandsontableOutput("parameters_metadata")
)
)
),
box(
width = 12,
title = "Covariates",
column(width = 6,
rHandsontableOutput("covariates_metadata")),
column(
width = 6,
uiOutput("metadata_categorical_covariates"),
rHandsontableOutput("selected_categorical_covariate_metadata"),
actionButton("update_categorical_levels", "Update categorical levels")
)
),
box(
width = 12,
title = "Unknown columns",
column(
6,
tags$p(
"Additional column(s) present in the dataset but absent of output tables:"
),
uiOutput("unknown_metadata")
),
column(
6,
wellPanel(
tags$label("Add column(s) selection to:"),
actionButton("add_to_parameters", "Individual parameters"),
actionButton("add_to_categorical_cov", "Categorical covariates"),
actionButton("add_to_continuous_cov", "Continuous covariates"),
actionButton("add_to_idv", "Independent variables")
)
)
)
)
),
tabItem(tabName = "population",
fluidRow(
box(
width = 8,
title = "Estimations",
uiOutput("estimation_selection")
),
box(
width = 4,
title = "Report",
radioButtons(
"estimation_report_format",
"Document format",
c("PDF", "HTML", "Word"),
inline = TRUE
),
fluidRow(column(
6,
downloadButton("estimation_download_report", "Download report")
),
column(
6, actionButton("estimation_save_report", "Save in run files")
))
),
tabBox(
width = 12,
tabPanel(
title = "Summary",
fluidRow(
valueBoxOutput(width = 4, "termination_box"),
valueBoxOutput(width = 4, "estimation_box"),
valueBoxOutput(width = 4, "ofv_box"),
valueBoxOutput(width = 4, "eigenratio_box"),
valueBoxOutput(width = 4, "correlation_box"),
valueBoxOutput(width = 4, "aic_box"),
valueBoxOutput(width = 4, "bic_box"),
valueBoxOutput(width = 4, "parallel_box")
),
h4("Termination messages"),
verbatimTextOutput("termination_messages"),
h4("Termination information"),
verbatimTextOutput("termination_information")
),
tabPanel(title = "Fixed effects (THETAs)",
extendedDT_with_code_UI("thetas_table")),
tabPanel(
title = "Random effects (OMEGA)",
h4("Covariances (ETA:ETA)"),
extendedDT_with_code_UI("omega_table"),
p("The coefficient of variation is given considering log-normal variability for parameters:",
withMathJax('$$CV(e^{\\eta})=\\sqrt{e^{\\omega^{2}}-1}$$')),
br(),
fluidRow(
column(
width = 6,
h4("Variance-Covariance Matrix"),
rHandsontableOutput("omega_matrix_table")
),
column(
width = 6,
h4("Correlation Matrix"),
rHandsontableOutput("correlation_matrix_table")
)
),
br(),
h4("ETA bars"),
extendedDT_with_code_UI("eta_bars_table")
),
tabPanel(title = "Residual error terms (SIGMA)",
fluidRow(
column(
width = 8,
h4("Covariances (EPS:EPS)"),
extendedDT_with_code_UI("sigma_table")
),
column(
width = 4,
h4("SIGMA matrix"),
rHandsontableOutput("sigma_matrix")
)
)),
tabPanel(
title = "Shrinkage",
h4("Random effects"),
extendedDT_with_code_UI("eta_shrinkage_table"),
br(),
h4("Residual error terms"),
extendedDT_with_code_UI("eps_shrinkage_table"),
br(),
h4("Legend"),
div(tags$ul(
tags$li(
"ETA shrinkage: Inter-subject shrinkage for each ETA",
withMathJax('$$\\eta_{shrink}=1-\\dfrac{SD(\\hat\\eta)}{\\omega}$$')
),
tags$li(
"EBV shrinkage: Shrinkage based on the average empirical Bayes variance"
),
tags$li(
"EPS shrinkage: Residual error shrinkage for each residual error",
withMathJax('$$\\epsilon_{shrink}=1-SD(IWRES)$$')
)
))
),
tabPanel(
title = "Convergence",
h4("Objective function"),
tabBox(
width = 12,
tabPanel(
"Population",
extendedPlotUI("ofv_convergence_plot", title = "ofv-convergence")
),
tabPanel(
"Individuals",
extendedPlotUI("individual_ofv_plot", title = "individual_ofv")
)
),
h4("Parameters convergence"),
selectInput(
"convergence_parameters",
"Population parameters",
choices = c(
"THETAs (Fixed effects)" = "theta",
"OMEGAs (Random effects covariances)" = "omega",
"SIGMAs (Residual error terms covariances)" = "sigma"
),
selected = "theta"
),
extendedPlotUI(
"pop_parameters_convergence_plot",
verbose = TRUE,
title = "parameters-convergence",
height = 700
)
)
)
)),
tabItem(tabName = "diagnostics",
fluidRow(
box(width = 2,
title = "Compartments",
uiOutput("cmt")),
box(
width = 10,
title = "Plots settings",
fluidRow(
column(2, selectInput(
"plots_x_scale",
"X-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
)),
column(2,
selectInput(
"plots_y_scale",
"Y-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
)),
column(
2,
selectInput(
"facet_scales",
"Scales",
choices = c(
"Free" = "free",
"Free X" = "free_x",
"Free Y" = "free_y",
"Fixed" = "fixed"
),
selected = "free"
)
),
column(
2,
conditionalPanel(
condition = "input.diagnostic_plots_box == 'dv_vs_pred_tab' | input.diagnostic_plots_box == 'residuals_tab'",
selectInput("diag_smoothing_method", "Smoothing method", choices = smoothing_methods)
)
),
column(
2,
conditionalPanel(
condition = "input.diagnostic_plots_box != 'residuals_tab_tab'",
checkboxInput("diag_log_dv", "Log-transformed DV", value = FALSE)
),
checkboxInput("diag_avoid_zero", "Exclude baseline", value = TRUE),
conditionalPanel(
condition = "input.diagnostic_plots_box != 'indiv_tab'",
checkboxInput("transparency", "Transparency", value = FALSE)
)
),
column(
2,
conditionalPanel(
condition = "input.diagnostic_plots_box != 'indiv_tab'",
selectizeInput(
"diagnostic_split_by",
"Split by",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates',
onInitialize = I('function() { this.setValue(""); }')
)
)
)
)
)
),
tabBox(
title = "Diagnostic plots",
width = 12,
id = "diagnostic_plots_box",
tabPanel(
value = "dv_vs_pred_tab",
title = "Dependent variables vs Predictions",
flowLayout(
uiOutput("dv_vs_pred_type"),
checkboxInput("stretch", label = "Stretch image", value = TRUE),
checkboxInput("dv_vs_pred_overlay", label = "Overlay prediction types", value = FALSE)
),
extendedPlotUI("dv_vs_pred_plot", title = "dv-vs-pred", height = 700),
h4("Selected points"),
DT::dataTableOutput("selected_dv_vs_pred_data")
),
tabPanel(
value = "spaghetti_tab",
title = "Spaghetti plot",
flowLayout(
uiOutput("spaghetti_idv"),
checkboxInput("spaghetti_split_facets", "Splits by facets", value = TRUE),
checkboxInput(
"spaghetti_split_means",
"Mean profiles (with respect to x-axis)",
value = FALSE
),
checkboxInput("spaghetti_mdv", "Show MDV", value = TRUE),
uiOutput("spaghetti_individuals")
),
extendedPlotUI("spaghetti_plot", title = "spaghetti", height = 700),
h4("Selected points"),
DT::dataTableOutput("selected_spaghetti_data")
),
tabPanel(
value = "residuals_tab",
title = "Residuals",
fluidRow(
column(3,
selectInput(
"residuals_plot_type",
"Plot type",
choices = c(
"Scatterplot" = "scatterplot",
"Histogram" = "histogram",
"Q-Q (Normal)" = "qq"
)
)),
column(6,
uiOutput("res_x_type"),
uiOutput("res_y_type")),
column(
3,
div(id = "res_histogram_widgets",
flowLayout(
sliderInput(
"residuals_hist_bins",
"Number of bins",
value = 30L,
min = 5L,
max = 50L
),
checkboxInput(
"residuals_histogram_empirical_density",
"Empirical distribution",
value = TRUE
),
checkboxInput(
"residuals_histogram_theoretical_distribution",
"Normal distribution",
value = TRUE
)
)),
div(id = "res_scatterplot_widgets",
fluidRow(
column(
6,
checkboxInput("res_absolute", "Absolute residuals", value = FALSE),
numericInput("res_y_limit", "Y-limit (abs)", value = NA)
),
column(
6,
checkboxInput("show_reference_value", "Show reference threshold", value = FALSE),
numericInput("res_reference_value", "Threshold value", value = 2)
)
))
)
),
extendedPlotUI("residuals_plot", title = "residuals", height = 700),
h4("Selected points"),
DT::dataTableOutput("selected_residuals_data")
),
tabPanel(
value = "indiv_tab",
title = "Individuals",
fluidRow(
column(
2,
uiOutput("indiv_idv"),
selectInput(
"individual_layout",
"Layout",
choices = c(
"1 individual" = 1,
"4 individuals" = 4,
"16 individuals" = 16
),
selected = 1
)
),
column(6,
uiOutput("individual_pages")),
column(
2,
uiOutput("indiv_pred_type"),
radioButtons(
"predictions_dots",
"Predictions display",
choices = c("Lines and dots" = TRUE, "Lines" = FALSE),
inline = TRUE
)
),
column(
2,
selectizeInput(
"indiv_cat_cov",
"Group by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
),
checkboxInput("show_observations", "Show observations", value = TRUE),
checkboxInput("show_indiv_mdv", "Show MDV", value = TRUE)
)
),
extendedPlotUI(
"individuals_plot",
title = "individuals",
height = 700,
verbose = TRUE
),
h4("Generate all subjects individual profiles"),
downloadButton("export_all_individual_profiles", "Export"),
actionButton("save_all_individual_profiles", "Save in run files")
)
)
)),
tabItem(tabName = "parameters_tab",
fluidRow(
box(
width = 12,
title = "Settings",
fluidRow(
column(
width = 2,
radioButtons(
"parameters_values_type",
"Individual parameters values",
choices = c("First value only" = 0, "All values" = 1),
inline = TRUE
)
),
column(
width = 2,
selectInput(
"individuals_parameters_type_selection",
"Parameters type",
choices = c(
"Random effects (ETAs)" = "eta",
"Individual parameters (POSTHOCs)" = "individual"
)
)
),
column(
width = 8,
selectizeInput(
"selected_parameters",
"Parameters selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select parameters...',
onInitialize = I('function() { this.setValue(""); }')
)
),
div(align = "center",
actionButton("parameters_refresh", "Refresh"))
)
)
)
),
fluidRow(tabBox(
width = 12,
tabPanel(
title = "Distributions",
h4("Visualization"),
flowLayout(
selectInput(
"parameters_distributions_plot_type",
"Plot type",
choices = c(
"Histogram" = "histogram",
"Density" = "density",
"Boxplot" = "boxplot",
"Q-Q (Normal)" = "qq"
)
),
conditionalPanel(
condition = "input.parameters_distributions_plot_type == 'histogram'",
sliderInput(
"parameters_hist_bins",
"Number of bins",
value = 30L,
min = 5L,
max = 50L
)
),
selectizeInput(
"parameters_distributions_split_by",
"Splitting options",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
),
div(
checkboxInput(
"overlay_parameters_distributions_splits",
"Overlay splits",
value = TRUE
),
conditionalPanel(
condition = "input.parameters_distributions_plot_type == 'boxplot'",
checkboxInput(
"parameters_distributions_boxplot_facetted",
"Facetted boxplot",
value = TRUE
),
checkboxInput(
"parameters_distributions_drop_unused_levels",
"Drop unused categorical levels",
value = FALSE
)
)
),
selectInput(
"parameters_distributions_facet_scales",
"Scales",
choices = c("Free XY" = "free", "Free X" = "free_x")
),
flowLayout(checkboxInput("parameters_histogram_empirical_density",
"Empirical density",
value = FALSE),
checkboxInput("parameters_histogram_theoretical_distribution",
"Normal distribution",
value = FALSE))
),
extendedPlotUI(
"parameters_distributions_plot",
title = "parameters-distributions",
height = 700
),
fluidRow(tabBox(
width = 12,
title = "Tables",
tabPanel(
"Summary statistics",
extendedDT_with_code_UI("parameters_distributions_summary")
),
tabPanel(
"Individual values",
extendedDTUI("parameters_individual_table")
)
))
),
tabPanel(
title = "Correlations",
fluidRow(
column(
width = 6,
flowLayout(
selectInput(
"parameters_correlations_method",
"Correlation method",
choices = c(
"pearson",
"kendall (heatmap only)" = "kendall",
"spearman (heatmap only)" = "spearman"
)
),
selectInput(
"parameters_correlations_plot_type",
"Plot type",
choices = c("Heatmap" = "heatmap", "Scatterplots" = "scatterplot")
),
conditionalPanel(
condition = "input.parameters_correlations_plot_type == 'scatterplot'",
selectInput(
"parameters_correlations_smooth",
"Smoothing method",
choices = smoothing_methods
)
)
),
extendedPlotUI(
"parameters_correlations_plot",
zoomable = FALSE,
verbose = TRUE,
title = "parameters-correlations",
export_width = 210,
export_height = 210,
height = 700,
click = "parameters_correlations_plot_click"
)
),
column(width = 6,
DT::dataTableOutput("parameters_correlations_table"))
),
h4("Selected correlation"),
fluidRow(
column(
width = 6,
flowLayout(
selectInput("p_p_correlations_smooth", "Smoothing method",
choices = smoothing_methods),
selectizeInput(
"p_p_correlations_group",
"Highlight by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Select categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
extendedPlotUI("p_p_correlation_plot",
title = "parameters-correlations", height = 700)
),
column(width = 6,
DT::dataTableOutput("selected_p_p_points_table"))
)
)
))),
tabItem(tabName = "covariates_tab",
fluidRow(
box(
width = 12,
title = "Settings",
radioButtons(
"covariates_values_type",
"Individual covariates values",
choices = c("First value only" = 0, "All values" = 1),
inline = TRUE
)
)
),
fluidRow(tabBox(
width = 12,
tabPanel(
title = "Continuous covariates",
selectizeInput(
"selected_continuous_covariates",
"Continuous covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
),
width = 1000
),
div(
align = "center",
actionButton("continuous_covariates_refresh", "Refresh")
),
fluidRow(tabBox(
width = 12,
tabPanel(
title = "Distributions",
h4("Visualization"),
flowLayout(
selectInput(
"continuous_covariates_distributions_plot_type",
"Plot type",
choices = c(
"Histogram" = "histogram",
"Density" = "density",
"Boxplot" = "boxplot",
"Q-Q (Normal)" = "qq"
)
),
conditionalPanel(
condition = "input.continuous_covariates_distributions_plot_type == 'histogram'",
sliderInput(
"covariates_hist_bins",
"Number of bins",
value = 30L,
min = 5L,
max = 50L
)
),
selectizeInput(
"continuous_covariates_distributions_split_by",
"Splitting options",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
),
div(
checkboxInput(
"overlay_continuous_covariates_distributions_splits",
"Overlay splits",
value = TRUE
)
),
selectInput(
"continuous_covariates_distributions_facet_scales",
"Scales",
choices = c("Free XY" = "free", "Free X" = "free_x")
)
),
extendedPlotUI(
"continuous_covariates_distributions_plot",
title = "covariates-distributions",
height = 700
),
fluidRow(tabBox(
width = 12,
title = "Tables",
tabPanel(
"Summary statistics",
extendedDTUI("continuous_covariates_distributions_summary")
),
tabPanel(
"Individual values",
extendedDTUI("continuous_covariates_individual_table")
)
))
),
tabPanel(
title = "Correlations",
fluidRow(
column(
width = 6,
flowLayout(
selectInput(
"covariates_correlations_method",
"Correlation method",
choices = c(
"pearson",
"kendall (heatmap only)" = "kendall",
"spearman (heatmap only)" = "spearman"
)
),
selectInput(
"covariates_correlations_plot_type",
"Plot type",
choices = c("Heatmap" = "heatmap", "Scatterplots" = "scatterplot")
),
conditionalPanel(
condition = "input.covariates_correlations_plot_type == 'scatterplot'",
selectInput(
"covariates_correlations_smooth",
"Smoothing method",
choices = smoothing_methods
)
)
),
extendedPlotUI(
"covariates_correlations_plot",
zoomable = FALSE,
verbose = TRUE,
title = "continuous-covariates-correlations",
export_width = 210,
export_height = 210,
height = 700,
click = "covariates_correlations_plot_click"
)
),
column(width = 6,
DT::dataTableOutput("covariates_correlations_table"))
),
h4("Selected correlation"),
fluidRow(
column(
width = 6,
flowLayout(
selectInput("c_c_correlations_smooth", "Smoothing method", choices = smoothing_methods),
selectizeInput(
"c_c_correlations_group",
"Highlight by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Select categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
extendedPlotUI("c_c_correlation_plot",
title = "continuous-covariates-correlation", height = 700)
),
column(width = 6,
DT::dataTableOutput("selected_c_c_points_table"))
)
)
))
),
tabPanel(
title = "Categorical covariates",
selectizeInput(
"selected_categorical_covariates",
"Categorical covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
),
width = 1000
),
div(
align = "center",
actionButton("categorical_covariates_refresh", "Refresh")
),
flowLayout(
selectInput(
"categorical_covariates_show_frequency",
"Y-axis",
choices = c("Count" = FALSE, "Frequency" = TRUE)
),
selectizeInput(
"categorical_covariates_distributions_split_by",
"Splitting options",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
),
checkboxInput("categorical_covariates_drop", "Drop empty levels", value = FALSE),
checkboxInput("categorical_covariates_order", "Order values", value = TRUE),
selectInput(
"categorical_covariates_bar_adjustment",
"Bar adjustment",
choices = c("dodge", "stack", "fill")
)
),
extendedPlotUI(
"categorical_covariates_distributions_plot",
title = "categorical-covariates-distributions",
height = 700
),
fluidRow(tabBox(
width = 12,
title = "Tables",
tabPanel(
"Summary statistics",
extendedDTUI("categorical_covariates_distributions_summary")
),
tabPanel(
"Individual values",
extendedDTUI("categorical_covariates_individual_table")
)
))
)
))),
tabItem(tabName = "params_vs_covs_tab",
fluidRow(
box(
width = 12,
title = "Settings",
fluidRow(
column(
width = 2,
radioButtons(
"p_c_values_type",
"Individual parameters and covariates values",
choices = c("First value only" = 0, "All values" = 1),
inline = TRUE
)
),
column(
width = 2,
selectInput(
"p_c_parameters_type_selection",
"Parameters type",
choices = c(
"Random effects (ETAs)" = "eta",
"Individual parameters (POSTHOCs)" = "individual"
)
)
),
column(
width = 8,
selectizeInput(
"p_c_selected_parameters",
"Parameters selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select parameters...',
onInitialize = I('function() { this.setValue(""); }')
)
),
conditionalPanel(
condition = "input.covariates_type_tab == 'continuous_tab'",
selectizeInput(
"p_c_selected_continuous_covariates",
"Continuous covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
conditionalPanel(
condition = "input.covariates_type_tab == 'categorical_tab'",
selectizeInput(
"p_c_selected_categorical_covariates",
"Categorical covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
div(
align = "center",
actionButton("parameters_vs_covariates_refresh", "Refresh")
)
)
)
),
tabBox(
title = "Covariates",
id = "covariates_type_tab",
width = 12,
tabPanel(
title = "Continuous",
value = "continuous_tab",
fluidRow(
column(
width = 6,
flowLayout(
selectInput(
"parameters_covariates_correlations_method",
"Correlation method",
choices = c("pearson",
"kendall",
"spearman")
)
),
extendedPlotUI(
"parameters_covariates_correlations_plot",
zoomable = FALSE,
verbose = TRUE,
title = "parameters-covariates-correlations",
export_width = 210,
export_height = 210,
height = 700,
click = "parameters_covariates_correlations_plot_click"
)
),
column(
width = 6,
DT::dataTableOutput("parameters_covariates_correlations_table")
)
),
h4("Selected correlation"),
fluidRow(
column(
width = 6,
flowLayout(
selectInput("p_c_correlations_smooth", "Smoothing method", choices = smoothing_methods),
selectizeInput(
"p_c_correlations_group",
"Highlight by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Select categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
extendedPlotUI("p_c_correlation_plot",
title = "parameters-covariates-correlations", height = 700)
),
column(width = 6,
DT::dataTableOutput("selected_p_c_points_table"))
)
),
tabPanel(
title = "Categorical",
value = "categorical_tab",
p(
"Select parameters and covariates and click the \"Refresh\" button"
),
extendedPlotUI(
"parameters_cat_covs_plot",
title = "parameters-vs-categorical-covariates",
height = 700
)
)
)
)),
tabItem(
tabName = "quality_criteria",
fluidRow(
box(
width = 8,
title = "Settings",
flowLayout(
uiOutput("qc_pred_type"),
checkboxInput("qc_log_data", "Log-transformed data", value = FALSE),
selectInput(
"qc_alpha",
"Confidence interval for tests",
choices = c(
"90%" = 0.10,
"95%" = 0.05,
"99%" = 0.01
),
selected = 0.05
),
selectizeInput(
"qc_split_by",
"Split by",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select compartment or categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
)
),
box(
width = 4,
title = "Report",
radioButtons(
"qc_report_format",
"Document format",
c("PDF", "HTML", "Word"),
inline = TRUE
),
fluidRow(column(
6, downloadButton("qc_download_report", "Download report")
),
column(
6, actionButton("qc_save_report", "Save in run files")
))
),
tabBox(
width = 12,
tabPanel(title = "Standard QC",
fluidRow(column(9,
DT::dataTableOutput("qc_standard"),
h4("Bias (Mean Prediction Error)"),
DT::dataTableOutput("qc_bias"),
h4("Precision (Root Mean Square Error)"),
DT::dataTableOutput("qc_precision")),
column(3,
h4("Formulas"),
tags$ul(
tags$li("Maximal Error",
withMathJax('$$ME=max(|pred_i-obs_i|)$$')),
tags$li(
"Absolute Average Fold Error",
withMathJax(
'$$AAFE=10^{\\dfrac{1}{N}{\\sum_{i=1}^N{|log(\\dfrac{pred_i}{obs_i}) |}}}$$'
)
),
tags$li("Mean Prediction Error",
withMathJax('$$MPE=\\dfrac{1}{N}{\\sum_{i=1}^N{pred_i-obs_i}}$$')),
tags$li("Mean Prediction Error (relative to the mean)",
withMathJax('$$MPE(\\%)=\\dfrac{MPE}{\\overline{obs}}$$')),
tags$li("Root Mean Square Error",
withMathJax('$$RMSE=\\sqrt{\\dfrac{\\sum_{i=1}^N{(pred_i-obs_i)^2}}{N}}$$')),
tags$li("RMSE (relative to the mean)",
withMathJax('$$RMSE(\\%)=\\dfrac{RMSE}{\\overline{obs}}$$'))
)))),
tabPanel(title = "Student's t-Tests",
em("Observations vs Predictions; paired, two-sided"),
DT::dataTableOutput("qc_t_test_obs"),
em("Residuals vs zero; paired, two-sided"),
DT::dataTableOutput("qc_t_test_res")),
tabPanel(title = "Correlation test",
em("Observations vs Predictions"),
DT::dataTableOutput("qc_corr_test")),
tabPanel(title = "Linear regression",
DT::dataTableOutput("qc_lin_reg"),
uiOutput("run_qc_lin_reg_plot")),
tabPanel(title = "R code",
shinyAce::aceEditor("qc_r_code", mode = "r", height = "100px", readOnly = TRUE))
)
)
),
tabItem(tabName = "outliers",
fluidRow(
box(
width = 8,
title = "Settings",
flowLayout(
uiOutput("outliers_cmt"),
uiOutput("outliers_restype"),
checkboxInput("outliers_avoid_zero", "Exclude baseline", value = TRUE),
conditionalPanel(
condition = "input.outliers_test_tab != 'boxplot'",
selectInput(
"outliers_pvalue",
"p-value threshold",
choices = c("1%" = 0.01, "5%" = 0.05, "10%" = 0.1),
selected = 0.05
)
)
)
),
box(
width = 4,
title = "Report",
radioButtons(
"outliers_report_format",
"Document format",
c("PDF", "HTML", "Word"),
inline = TRUE
),
fluidRow(column(
6,
downloadButton("outliers_download_report", "Download report")
),
column(
6, actionButton("outliers_save_report", "Save in run files")
))
),
tabBox(
width = 12,
id = "outliers_test_tab",
title = "Procedures",
tabPanel(title = "Normality check",
fluidRow(
column(
width = 8,
h4("Q-Q (Normal) plot"),
div(align="center",
extendedPlotUI(
"outliers_qqplot",
title = "outliers-qqplot",
width = 500,
height = 500,
export_width = 210,
export_height = 210
)
)
),
column(
width = 4,
h4("Kolmogorov-Smirnov test"),
h6("Note: duplicate values are considered as one unique value"),
DT::dataTableOutput("outliers_ks_table"),
br(),
verbatimTextOutput("outliers_ks_decision")
)
),
fluidRow(column(
12,
h4("Selected residuals"),
DT::dataTableOutput("selected_outliers_residuals_table")
))),
tabPanel(
title = "Grubb's test",
h4("Outliers"),
extendedDTUI("outliers_grubbs_table"),
br(),
verbatimTextOutput("outliers_grubbs_message"),
h6(
"Note: Grubb's test makes the assumption that data is normally distributed."
)
),
tabPanel(
title = "Non-parametric",
value = "boxplot",
flowLayout(
numericInput(
"outliers_k_coefficient",
"k coefficient (default = 3)",
value = 3,
min = 0,
step = 0.1
)
),
fluidRow(
column(
width = 6,
h4("Outliers"),
extendedDTUI("outliers_boxplot_table"),
br(),
verbatimTextOutput("outliers_boxplot_message")
),
column(
width = 6,
h4("Box-and-whisker plot"),
extendedPlotUI(
"outliers_boxplot",
title = "outliers-boxplot",
height = 600,
export_width = 210,
export_height = 210
)
)
),
h4("Note"),
div(
"Observations are detected as outliers if they are outside of the following interval:",
withMathJax(
"$$\\mathopen{[}F_f - k \\cdot IQR\\,;F_t + k \\cdot IQR\\mathclose{]}$$"
),
"with Ff being the first quartile, Ft the third quartile, IQR the inter-quartile range (Ft-Ff) and k is a coefficient (length of the whiskers as multiple of IQR)"
)
)
)
)),
tabItem(tabName = "vpc",
fluidRow(
box(
width = 12,
title = "Visual Predictive Checks",
tags$div(
"Based on the run estimation results, a control stream is generated to perform simulations for Visual Predictive Checks."
),
br(),
tags$strong("The following must be performed to generate VPC:"),
tags$ol(
tags$li("Set the local paths to NONMEM and NMTRAN check executables"),
tags$li("Set number of simulations and seed"),
tags$li("Generate the control stream for simulation"),
tags$li(
"Run simulations locally using NONMEM (do not close the application until computation is finished)"
)
),
br(),
tags$div("The VPC plots and data are then computed by the ", tags$strong("vpc"),
" R package (docs: ", tags$a(href = "http://vpc.ronkeizer.com/", target = "_blank", "http://vpc.ronkeizer.com/"), ")")
,
fluidRow(column(
8,
h4("NONMEM paths"),
fluidRow(column(6, uiOutput("nm_exe")),
column(6, uiOutput("nmcheck_exe")))
),
column(
4,
h4("Simulations settings"),
fluidRow(column(
6,
numericInput(
"vpc_n_simulations",
"Number of simulations",
value = 100,
min = 1,
width = 175
)
),
column(6,numericInput("vpc_seed", "Simulations seed", value = 123456, width = 175)
))
)),
tags$div(
align = "center",
actionButton("generate_control_stream", "Generate control stream")
),
uiOutput("vpc_location")
),
tabBox(
width = 12,
title = "VPC results",
tabPanel("Visualization",
fluidRow(
column(3,
fluidRow(
column(
6,
uiOutput("vpc_idv"),
uiOutput("vpc_cmt"),
checkboxInput("vpc_correction", "Prediction-correction", value = FALSE),
sliderInput(
"vpc_pi",
"Prediction interval (PI) range",
value = c(0.05, 0.95),
min = 0,
max = 1,
step = 0.01
),
sliderInput(
"vpc_ci",
"Confidence interval (CI) range",
value = c(0.05, 0.95),
min = 0,
max = 1,
step = 0.01
),
selectInput("vpc_bin_mid", "Bin mid", choices = c("mean", "middle")),
selectInput(
"vpc_binning",
"Binning method",
choices = c(
"data (default)" = "data",
"time" = "time",
"auto",
"density",
"jenks",
"none",
"pretty"
),
selected = "data"
),
numericInput("vpc_n_bins", "Number of bins", value = NA)
),
column(
6,
tags$label("Plot:"),
checkboxInput("vpc_obs_dv", "Observations", value = TRUE),
checkboxInput("vpc_obs_ci", "Observations CI", value = TRUE),
checkboxInput("vpc_show_pi", "Prediction interval", value = FALSE),
checkboxInput("vpc_pi_as_area", "PI as area", value = FALSE),
checkboxInput("vpc_pi_ci", "CI around PI", value = TRUE),
checkboxInput("vpc_obs_median", "Observations median", value = TRUE),
checkboxInput("vpc_sim_median", "Simulations median", value = FALSE),
checkboxInput("vpc_sim_median_ci", "Simulations median CI", value = TRUE),
checkboxInput("vpc_smooth", "Smooth", value = TRUE),
numericInput("vpc_lloq", "Lower limit of quantification", value = NA),
selectInput(
"vpc_x_scale",
"X-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
),
selectInput(
"vpc_y_scale",
"Y-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
),
selectInput(
"vpc_facet_scale",
"Facet scales",
choices = c(
"Free" = "free",
"Free X" = "free_x",
"Free Y" = "free_y",
"Fixed" = "fixed"
),
selected = "fixed"
)
)
)),
column(
9,
fluidRow(column(2, uiOutput("vpc_strat1")), column(10, uiOutput(
"vpc_strat1_levels"
))),
fluidRow(column(2, uiOutput("vpc_strat2")), column(10, uiOutput(
"vpc_strat2_levels"
))),
extendedPlotUI("vpc_plot", title = "VPC", height = 700)
)
)),
tabPanel("Data",
fluidRow(
column(12,
extendedDTUI("vpc_db_table", title = "VPC summary data"),
extendedDTUI("vpc_obs_table", title = "Observations"),
extendedDTUI("vpc_sim_table", title = "VPC Simulations"))
))
)
)),
tabItem(tabName = "comparison",
fluidRow(
box(
width = 12,
title = "Compare run to...",
fluidRow(
column(
2,
actionButton("select_runs_to_compare", "Select runs"),
actionButton("clear_comparison_table", "Clear selection")
),
column(
10,
flowLayout(
uiOutput("comparison_estimation_n"),
uiOutput("comparison_estimation_status")
),
uiOutput("comparison_estimation_method")
)
)
),
div(
id = "comparison-content",
tabBox(
width = 12,
title = "Comparison",
tabPanel(
width = 12,
"Summary table",
h4("Termination status"),
DT::dataTableOutput("comparison_status_summary"),
br(),
h4("Detailed comparison"),
fluidRow(column(
10,
checkboxGroupInput(
"comparison_details",
"Details",
choices = c(
"Info" = "INFO",
"Summary" = "SUMMARY",
"THETA" = "THETA",
"OMEGA" = "OMEGA",
"SIGMA" = "SIGMA",
"ETABAR" = "ETABAR",
"SHRINKAGE" = "SHRINKAGE",
"Filenames" = "FILES"
),
selected = c("SUMMARY", "THETA", "OMEGA"),
inline = TRUE
)
),
column(
2,
downloadButton("export_comparison_table", "Export")
)),
extendedDTUI("comparison_table")
),
tabPanel("Plots",
fluidRow(
tabBox(
width = 12,
tabPanel(
"Parameters",
fluidRow(
column(
2,
selectInput(
"comparison_distribution_plot",
label = "Plot type",
choices = c(
"Histogram" = "histogram",
"Density" = "density",
"Boxplot" = "boxplot"
)
)
),
column(
2,
radioButtons(
"comparison_plots_details",
"Parameters",
choices = c(
"THETA" = "THETA",
"OMEGA" = "OMEGA",
"SIGMA" = "SIGMA"
),
selected = "THETA",
inline = TRUE
)
),
column(8, uiOutput("comparison_plots_parameters"))
),
extendedPlotUI("comparison_param_plot", height = 700)
),
tabPanel("OFV",
extendedPlotUI("comparison_ofv_plot", height = 700))
))),
tabPanel(
"Statistics",
checkboxGroupInput(
"comparison_stats_table_detail",
"Details selection",
choices = c("THETA", "OMEGA", "SIGMA", "ETABAR", "SHRINKAGE"),
selected = c("THETA", "OMEGA", "SIGMA"),
inline = TRUE
),
checkboxInput("comparison_stats_table_rse", "RSE", value = FALSE),
extendedDTUI("comparison_stats_table")
)
)
)
)),
tabItem(tabName = "table_files",
fluidRow(
box(
width = 12,
title = "Output tables",
uiOutput("run_table_selection"),
checkboxInput("subject_first_row", "Subject first row only", value = FALSE),
extendedDTUI("run_table_output")
)
)),
tabItem(tabName = "control_stream_file",
fluidRow(
tabBox(
width = 12,
title = "Control stream",
tabPanel(title = "Raw",
verbatimTextOutput("control_stream_raw_content")),
tabPanel(title = "PMXploit control stream object",
verbatimTextOutput("control_stream_parsed_content"))
)
)),
tabItem(tabName = "report_file",
fluidRow(
box(
width = 12,
title = "Report file (NONMEM output)",
verbatimTextOutput("report_content")
)
))
)
)
)
pnolain/popkinr documentation built on Jan. 31, 2024, 7:05 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
suppressPackageStartupMessages({
library(shiny)
library(shinydashboard)
library(DT)
library(shinyjs)
library(shinyAce)
library(shinytoastr)
library(rhandsontable)
library(ggplot2)
library(tidyr)
library(dplyr)
library(stringr)
library(purrr)
library(readr)
library(lubridate)
library(hms)
library(broom)
library(future)
library(pmxploit)
library(rlang)
library(xml2)
library(vpc)
library(shinycssloaders)
library(shinyTree)
library(jsonlite)
})
source("modules/extendedPlot.R")
source("modules/extendedDT.R")
source("modules/extendedDT_with_code.R")
# Show VPC tab only if a NONMEM executable path is available
vpc_tab <- (if(Sys.getenv("NM_EXE") != "") {
menuItem("VPC", tabName = "vpc", icon = icon("area-chart"))
} else {
NULL
})
app_title <- "PopkinR - NONMEM Toolbox"
smoothing_methods <- c(
"None" = "none",
"Linear model" = "lm",
"Local Polynomial Regression" = "loess"
)
# Edit header bar to integrate run path
db_header <-
dashboardHeader(title = app_title, titleWidth = 400)
db_header$children[[3]]$children[[3]] <-
uiOutput("application_top_text")
dashboardPage(
skin = "green",
title = app_title,
db_header,
dashboardSidebar(
sidebarMenu(
id = "main_menu",
menuItem(
"Home",
tabName = "home",
icon = icon("home"),
selected = TRUE
),
menuItem("Metadata", tabName = "metadata", icon = icon("info")),
menuItem("Population", tabName = "population", icon = icon("globe")),
menuItem("Diagnostics", tabName = "diagnostics", icon = icon("search")),
menuItem(
"Individuals",
icon = icon("users"),
menuSubItem("Parameters", tabName = "parameters_tab", icon = icon("cogs")),
menuSubItem(
"Covariates",
tabName = "covariates_tab",
icon = icon("pie-chart")
),
menuSubItem(
"Parameters vs Covariates",
tabName = "params_vs_covs_tab",
icon = icon("table")
)
),
menuItem(
"Quality criteria",
tabName = "quality_criteria",
icon = icon("check")
),
menuItem(
"Outliers detection",
tabName = "outliers",
icon = icon("user-secret")
),
vpc_tab,
menuItem(
"Compare to...",
tabName = "comparison",
icon = icon("sort-numeric-asc")
),
menuItem(
"Files",
icon = icon("copy"),
menuSubItem(
"Control stream",
tabName = "control_stream_file",
icon = icon("file-code")
),
menuSubItem("Report", tabName = "report_file", icon = icon("file-text")),
menuSubItem("Tables", tabName = "table_files", icon = icon("database"))
),
br(),
div(align = "center", actionButton("show_filters", "Filter data")),
hr(),
div(
align = "center",
img(
src = "media/pmxploit.svg",
width = 175,
height = 175
),
h6(
sprintf(
"PopkinR version: %s (%s)",
packageDescription("popkinr")$Version,
packageDescription("popkinr")$Date
)
),
h6("Patrick Nolain"),
tags$a(icon("github"), "github.com/pnolain", href="https://github.com/pnolain"),
hr(),
actionLink("example", "Demo")
)
)
),
dashboardBody(
useShinyjs(),
includeScript("www/js/sweetalert2.min.js"),
shiny::includeCSS("www/css/sweetalert2.min.css"),
includeScript("www/js/script.js"),
shiny::includeCSS("www/css/styles.css"),
useToastr(),
div(
id = "loading-content",
img(
src = "media/pmxploit_animated.svg",
width = 200,
height = 200
)
),
tabItems(
tabItem(tabName = "home",
fluidRow(
box(
id = "browsing_box",
width = 12,
title = "Select a run",
strong(actionLink("select_run", "Browse..."))
),
box(
id = "recent_box",
title = "History",
collapsible = TRUE,
width = 12,
uiOutput("open_last_run")
),
shinyjs::hidden(
div(id = "summary_box",
box(
title = "Summary",
width = 12,
h4("Estimation"),
fluidRow(
column(12,
DT::dataTableOutput("estimations_table"))
),
h4("Run information"),
fluidRow(
column(7,
uiOutput("run_information")),
column(
5,
fluidRow(
valueBoxOutput(width = 6, "subjects_box"),
valueBoxOutput(width = 6, "observations_box")
),
fluidRow(
valueBoxOutput(width = 6, "duration_box"),
valueBoxOutput(width = 6, "cpu_box")
),
fluidRow(valueBoxOutput(width = 6, "nm_version_box"))
)
)
))
)
)),
tabItem(
tabName = "metadata",
title = "Metadata",
fluidRow(
box(
width = 12,
title = "Metadata",
fluidRow(column(6,
tags$div(
tags$p(
"Metadata refer to details of the model allowing the user to manipulate several elements for plotting and reporting (e.g. compartments/parameters/covariates names, covariates types) by editing the following boxes."
),
tags$p("One can also:"),
tags$ul(
tags$li(
actionButton("browse_metadata", label = "Import"),
"metadata from another run"
),
tags$li(
actionButton("clear_metadata", label = "Clear"),
"(remove) metadata"
),
tags$li(
actionButton("reset_metadata", label = "Reset"),
"recent changes in metadata to their state at loading"
),
tags$li(
actionButton("save_metadata", label = "Save"),
"metadata to the run location"
)
)
)),
column(6, p()))
),
box(
width = 6,
title = "Compartments",
rHandsontableOutput("compartments_metadata")
),
box(
width = 6,
title = "Independent variables",
rHandsontableOutput("idv_metadata")
),
box(
width = 12,
title = "Parameters",
fluidRow(
width = 12,
column(
width = 4,
h4("Fixed effects (THETAs)"),
rHandsontableOutput("thetas_metadata")
),
column(
width = 4,
h4("Random effects (ETAs)"),
rHandsontableOutput("etas_metadata")
),
column(
width = 4,
h4("Individual parameters (POSTHOCs)"),
rHandsontableOutput("parameters_metadata")
)
)
),
box(
width = 12,
title = "Covariates",
column(width = 6,
rHandsontableOutput("covariates_metadata")),
column(
width = 6,
uiOutput("metadata_categorical_covariates"),
rHandsontableOutput("selected_categorical_covariate_metadata"),
actionButton("update_categorical_levels", "Update categorical levels")
)
),
box(
width = 12,
title = "Unknown columns",
column(
6,
tags$p(
"Additional column(s) present in the dataset but absent of output tables:"
),
uiOutput("unknown_metadata")
),
column(
6,
wellPanel(
tags$label("Add column(s) selection to:"),
actionButton("add_to_parameters", "Individual parameters"),
actionButton("add_to_categorical_cov", "Categorical covariates"),
actionButton("add_to_continuous_cov", "Continuous covariates"),
actionButton("add_to_idv", "Independent variables")
)
)
)
)
),
tabItem(tabName = "population",
fluidRow(
box(
width = 8,
title = "Estimations",
uiOutput("estimation_selection")
),
box(
width = 4,
title = "Report",
radioButtons(
"estimation_report_format",
"Document format",
c("PDF", "HTML", "Word"),
inline = TRUE
),
fluidRow(column(
6,
downloadButton("estimation_download_report", "Download report")
),
column(
6, actionButton("estimation_save_report", "Save in run files")
))
),
tabBox(
width = 12,
tabPanel(
title = "Summary",
fluidRow(
valueBoxOutput(width = 4, "termination_box"),
valueBoxOutput(width = 4, "estimation_box"),
valueBoxOutput(width = 4, "ofv_box"),
valueBoxOutput(width = 4, "eigenratio_box"),
valueBoxOutput(width = 4, "correlation_box"),
valueBoxOutput(width = 4, "aic_box"),
valueBoxOutput(width = 4, "bic_box"),
valueBoxOutput(width = 4, "parallel_box")
),
h4("Termination messages"),
verbatimTextOutput("termination_messages"),
h4("Termination information"),
verbatimTextOutput("termination_information")
),
tabPanel(title = "Fixed effects (THETAs)",
extendedDT_with_code_UI("thetas_table")),
tabPanel(
title = "Random effects (OMEGA)",
h4("Covariances (ETA:ETA)"),
extendedDT_with_code_UI("omega_table"),
p("The coefficient of variation is given considering log-normal variability for parameters:",
withMathJax('$$CV(e^{\\eta})=\\sqrt{e^{\\omega^{2}}-1}$$')),
br(),
fluidRow(
column(
width = 6,
h4("Variance-Covariance Matrix"),
rHandsontableOutput("omega_matrix_table")
),
column(
width = 6,
h4("Correlation Matrix"),
rHandsontableOutput("correlation_matrix_table")
)
),
br(),
h4("ETA bars"),
extendedDT_with_code_UI("eta_bars_table")
),
tabPanel(title = "Residual error terms (SIGMA)",
fluidRow(
column(
width = 8,
h4("Covariances (EPS:EPS)"),
extendedDT_with_code_UI("sigma_table")
),
column(
width = 4,
h4("SIGMA matrix"),
rHandsontableOutput("sigma_matrix")
)
)),
tabPanel(
title = "Shrinkage",
h4("Random effects"),
extendedDT_with_code_UI("eta_shrinkage_table"),
br(),
h4("Residual error terms"),
extendedDT_with_code_UI("eps_shrinkage_table"),
br(),
h4("Legend"),
div(tags$ul(
tags$li(
"ETA shrinkage: Inter-subject shrinkage for each ETA",
withMathJax('$$\\eta_{shrink}=1-\\dfrac{SD(\\hat\\eta)}{\\omega}$$')
),
tags$li(
"EBV shrinkage: Shrinkage based on the average empirical Bayes variance"
),
tags$li(
"EPS shrinkage: Residual error shrinkage for each residual error",
withMathJax('$$\\epsilon_{shrink}=1-SD(IWRES)$$')
)
))
),
tabPanel(
title = "Convergence",
h4("Objective function"),
tabBox(
width = 12,
tabPanel(
"Population",
extendedPlotUI("ofv_convergence_plot", title = "ofv-convergence")
),
tabPanel(
"Individuals",
extendedPlotUI("individual_ofv_plot", title = "individual_ofv")
)
),
h4("Parameters convergence"),
selectInput(
"convergence_parameters",
"Population parameters",
choices = c(
"THETAs (Fixed effects)" = "theta",
"OMEGAs (Random effects covariances)" = "omega",
"SIGMAs (Residual error terms covariances)" = "sigma"
),
selected = "theta"
),
extendedPlotUI(
"pop_parameters_convergence_plot",
verbose = TRUE,
title = "parameters-convergence",
height = 700
)
)
)
)),
tabItem(tabName = "diagnostics",
fluidRow(
box(width = 2,
title = "Compartments",
uiOutput("cmt")),
box(
width = 10,
title = "Plots settings",
fluidRow(
column(2, selectInput(
"plots_x_scale",
"X-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
)),
column(2,
selectInput(
"plots_y_scale",
"Y-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
)),
column(
2,
selectInput(
"facet_scales",
"Scales",
choices = c(
"Free" = "free",
"Free X" = "free_x",
"Free Y" = "free_y",
"Fixed" = "fixed"
),
selected = "free"
)
),
column(
2,
conditionalPanel(
condition = "input.diagnostic_plots_box == 'dv_vs_pred_tab' | input.diagnostic_plots_box == 'residuals_tab'",
selectInput("diag_smoothing_method", "Smoothing method", choices = smoothing_methods)
)
),
column(
2,
conditionalPanel(
condition = "input.diagnostic_plots_box != 'residuals_tab_tab'",
checkboxInput("diag_log_dv", "Log-transformed DV", value = FALSE)
),
checkboxInput("diag_avoid_zero", "Exclude baseline", value = TRUE),
conditionalPanel(
condition = "input.diagnostic_plots_box != 'indiv_tab'",
checkboxInput("transparency", "Transparency", value = FALSE)
)
),
column(
2,
conditionalPanel(
condition = "input.diagnostic_plots_box != 'indiv_tab'",
selectizeInput(
"diagnostic_split_by",
"Split by",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates',
onInitialize = I('function() { this.setValue(""); }')
)
)
)
)
)
),
tabBox(
title = "Diagnostic plots",
width = 12,
id = "diagnostic_plots_box",
tabPanel(
value = "dv_vs_pred_tab",
title = "Dependent variables vs Predictions",
flowLayout(
uiOutput("dv_vs_pred_type"),
checkboxInput("stretch", label = "Stretch image", value = TRUE),
checkboxInput("dv_vs_pred_overlay", label = "Overlay prediction types", value = FALSE)
),
extendedPlotUI("dv_vs_pred_plot", title = "dv-vs-pred", height = 700),
h4("Selected points"),
DT::dataTableOutput("selected_dv_vs_pred_data")
),
tabPanel(
value = "spaghetti_tab",
title = "Spaghetti plot",
flowLayout(
uiOutput("spaghetti_idv"),
checkboxInput("spaghetti_split_facets", "Splits by facets", value = TRUE),
checkboxInput(
"spaghetti_split_means",
"Mean profiles (with respect to x-axis)",
value = FALSE
),
checkboxInput("spaghetti_mdv", "Show MDV", value = TRUE),
uiOutput("spaghetti_individuals")
),
extendedPlotUI("spaghetti_plot", title = "spaghetti", height = 700),
h4("Selected points"),
DT::dataTableOutput("selected_spaghetti_data")
),
tabPanel(
value = "residuals_tab",
title = "Residuals",
fluidRow(
column(3,
selectInput(
"residuals_plot_type",
"Plot type",
choices = c(
"Scatterplot" = "scatterplot",
"Histogram" = "histogram",
"Q-Q (Normal)" = "qq"
)
)),
column(6,
uiOutput("res_x_type"),
uiOutput("res_y_type")),
column(
3,
div(id = "res_histogram_widgets",
flowLayout(
sliderInput(
"residuals_hist_bins",
"Number of bins",
value = 30L,
min = 5L,
max = 50L
),
checkboxInput(
"residuals_histogram_empirical_density",
"Empirical distribution",
value = TRUE
),
checkboxInput(
"residuals_histogram_theoretical_distribution",
"Normal distribution",
value = TRUE
)
)),
div(id = "res_scatterplot_widgets",
fluidRow(
column(
6,
checkboxInput("res_absolute", "Absolute residuals", value = FALSE),
numericInput("res_y_limit", "Y-limit (abs)", value = NA)
),
column(
6,
checkboxInput("show_reference_value", "Show reference threshold", value = FALSE),
numericInput("res_reference_value", "Threshold value", value = 2)
)
))
)
),
extendedPlotUI("residuals_plot", title = "residuals", height = 700),
h4("Selected points"),
DT::dataTableOutput("selected_residuals_data")
),
tabPanel(
value = "indiv_tab",
title = "Individuals",
fluidRow(
column(
2,
uiOutput("indiv_idv"),
selectInput(
"individual_layout",
"Layout",
choices = c(
"1 individual" = 1,
"4 individuals" = 4,
"16 individuals" = 16
),
selected = 1
)
),
column(6,
uiOutput("individual_pages")),
column(
2,
uiOutput("indiv_pred_type"),
radioButtons(
"predictions_dots",
"Predictions display",
choices = c("Lines and dots" = TRUE, "Lines" = FALSE),
inline = TRUE
)
),
column(
2,
selectizeInput(
"indiv_cat_cov",
"Group by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
),
checkboxInput("show_observations", "Show observations", value = TRUE),
checkboxInput("show_indiv_mdv", "Show MDV", value = TRUE)
)
),
extendedPlotUI(
"individuals_plot",
title = "individuals",
height = 700,
verbose = TRUE
),
h4("Generate all subjects individual profiles"),
downloadButton("export_all_individual_profiles", "Export"),
actionButton("save_all_individual_profiles", "Save in run files")
)
)
)),
tabItem(tabName = "parameters_tab",
fluidRow(
box(
width = 12,
title = "Settings",
fluidRow(
column(
width = 2,
radioButtons(
"parameters_values_type",
"Individual parameters values",
choices = c("First value only" = 0, "All values" = 1),
inline = TRUE
)
),
column(
width = 2,
selectInput(
"individuals_parameters_type_selection",
"Parameters type",
choices = c(
"Random effects (ETAs)" = "eta",
"Individual parameters (POSTHOCs)" = "individual"
)
)
),
column(
width = 8,
selectizeInput(
"selected_parameters",
"Parameters selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select parameters...',
onInitialize = I('function() { this.setValue(""); }')
)
),
div(align = "center",
actionButton("parameters_refresh", "Refresh"))
)
)
)
),
fluidRow(tabBox(
width = 12,
tabPanel(
title = "Distributions",
h4("Visualization"),
flowLayout(
selectInput(
"parameters_distributions_plot_type",
"Plot type",
choices = c(
"Histogram" = "histogram",
"Density" = "density",
"Boxplot" = "boxplot",
"Q-Q (Normal)" = "qq"
)
),
conditionalPanel(
condition = "input.parameters_distributions_plot_type == 'histogram'",
sliderInput(
"parameters_hist_bins",
"Number of bins",
value = 30L,
min = 5L,
max = 50L
)
),
selectizeInput(
"parameters_distributions_split_by",
"Splitting options",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
),
div(
checkboxInput(
"overlay_parameters_distributions_splits",
"Overlay splits",
value = TRUE
),
conditionalPanel(
condition = "input.parameters_distributions_plot_type == 'boxplot'",
checkboxInput(
"parameters_distributions_boxplot_facetted",
"Facetted boxplot",
value = TRUE
),
checkboxInput(
"parameters_distributions_drop_unused_levels",
"Drop unused categorical levels",
value = FALSE
)
)
),
selectInput(
"parameters_distributions_facet_scales",
"Scales",
choices = c("Free XY" = "free", "Free X" = "free_x")
),
flowLayout(checkboxInput("parameters_histogram_empirical_density",
"Empirical density",
value = FALSE),
checkboxInput("parameters_histogram_theoretical_distribution",
"Normal distribution",
value = FALSE))
),
extendedPlotUI(
"parameters_distributions_plot",
title = "parameters-distributions",
height = 700
),
fluidRow(tabBox(
width = 12,
title = "Tables",
tabPanel(
"Summary statistics",
extendedDT_with_code_UI("parameters_distributions_summary")
),
tabPanel(
"Individual values",
extendedDTUI("parameters_individual_table")
)
))
),
tabPanel(
title = "Correlations",
fluidRow(
column(
width = 6,
flowLayout(
selectInput(
"parameters_correlations_method",
"Correlation method",
choices = c(
"pearson",
"kendall (heatmap only)" = "kendall",
"spearman (heatmap only)" = "spearman"
)
),
selectInput(
"parameters_correlations_plot_type",
"Plot type",
choices = c("Heatmap" = "heatmap", "Scatterplots" = "scatterplot")
),
conditionalPanel(
condition = "input.parameters_correlations_plot_type == 'scatterplot'",
selectInput(
"parameters_correlations_smooth",
"Smoothing method",
choices = smoothing_methods
)
)
),
extendedPlotUI(
"parameters_correlations_plot",
zoomable = FALSE,
verbose = TRUE,
title = "parameters-correlations",
export_width = 210,
export_height = 210,
height = 700,
click = "parameters_correlations_plot_click"
)
),
column(width = 6,
DT::dataTableOutput("parameters_correlations_table"))
),
h4("Selected correlation"),
fluidRow(
column(
width = 6,
flowLayout(
selectInput("p_p_correlations_smooth", "Smoothing method",
choices = smoothing_methods),
selectizeInput(
"p_p_correlations_group",
"Highlight by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Select categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
extendedPlotUI("p_p_correlation_plot",
title = "parameters-correlations", height = 700)
),
column(width = 6,
DT::dataTableOutput("selected_p_p_points_table"))
)
)
))),
tabItem(tabName = "covariates_tab",
fluidRow(
box(
width = 12,
title = "Settings",
radioButtons(
"covariates_values_type",
"Individual covariates values",
choices = c("First value only" = 0, "All values" = 1),
inline = TRUE
)
)
),
fluidRow(tabBox(
width = 12,
tabPanel(
title = "Continuous covariates",
selectizeInput(
"selected_continuous_covariates",
"Continuous covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
),
width = 1000
),
div(
align = "center",
actionButton("continuous_covariates_refresh", "Refresh")
),
fluidRow(tabBox(
width = 12,
tabPanel(
title = "Distributions",
h4("Visualization"),
flowLayout(
selectInput(
"continuous_covariates_distributions_plot_type",
"Plot type",
choices = c(
"Histogram" = "histogram",
"Density" = "density",
"Boxplot" = "boxplot",
"Q-Q (Normal)" = "qq"
)
),
conditionalPanel(
condition = "input.continuous_covariates_distributions_plot_type == 'histogram'",
sliderInput(
"covariates_hist_bins",
"Number of bins",
value = 30L,
min = 5L,
max = 50L
)
),
selectizeInput(
"continuous_covariates_distributions_split_by",
"Splitting options",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
),
div(
checkboxInput(
"overlay_continuous_covariates_distributions_splits",
"Overlay splits",
value = TRUE
)
),
selectInput(
"continuous_covariates_distributions_facet_scales",
"Scales",
choices = c("Free XY" = "free", "Free X" = "free_x")
)
),
extendedPlotUI(
"continuous_covariates_distributions_plot",
title = "covariates-distributions",
height = 700
),
fluidRow(tabBox(
width = 12,
title = "Tables",
tabPanel(
"Summary statistics",
extendedDTUI("continuous_covariates_distributions_summary")
),
tabPanel(
"Individual values",
extendedDTUI("continuous_covariates_individual_table")
)
))
),
tabPanel(
title = "Correlations",
fluidRow(
column(
width = 6,
flowLayout(
selectInput(
"covariates_correlations_method",
"Correlation method",
choices = c(
"pearson",
"kendall (heatmap only)" = "kendall",
"spearman (heatmap only)" = "spearman"
)
),
selectInput(
"covariates_correlations_plot_type",
"Plot type",
choices = c("Heatmap" = "heatmap", "Scatterplots" = "scatterplot")
),
conditionalPanel(
condition = "input.covariates_correlations_plot_type == 'scatterplot'",
selectInput(
"covariates_correlations_smooth",
"Smoothing method",
choices = smoothing_methods
)
)
),
extendedPlotUI(
"covariates_correlations_plot",
zoomable = FALSE,
verbose = TRUE,
title = "continuous-covariates-correlations",
export_width = 210,
export_height = 210,
height = 700,
click = "covariates_correlations_plot_click"
)
),
column(width = 6,
DT::dataTableOutput("covariates_correlations_table"))
),
h4("Selected correlation"),
fluidRow(
column(
width = 6,
flowLayout(
selectInput("c_c_correlations_smooth", "Smoothing method", choices = smoothing_methods),
selectizeInput(
"c_c_correlations_group",
"Highlight by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Select categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
extendedPlotUI("c_c_correlation_plot",
title = "continuous-covariates-correlation", height = 700)
),
column(width = 6,
DT::dataTableOutput("selected_c_c_points_table"))
)
)
))
),
tabPanel(
title = "Categorical covariates",
selectizeInput(
"selected_categorical_covariates",
"Categorical covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
),
width = 1000
),
div(
align = "center",
actionButton("categorical_covariates_refresh", "Refresh")
),
flowLayout(
selectInput(
"categorical_covariates_show_frequency",
"Y-axis",
choices = c("Count" = FALSE, "Frequency" = TRUE)
),
selectizeInput(
"categorical_covariates_distributions_split_by",
"Splitting options",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select categorical covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
),
checkboxInput("categorical_covariates_drop", "Drop empty levels", value = FALSE),
checkboxInput("categorical_covariates_order", "Order values", value = TRUE),
selectInput(
"categorical_covariates_bar_adjustment",
"Bar adjustment",
choices = c("dodge", "stack", "fill")
)
),
extendedPlotUI(
"categorical_covariates_distributions_plot",
title = "categorical-covariates-distributions",
height = 700
),
fluidRow(tabBox(
width = 12,
title = "Tables",
tabPanel(
"Summary statistics",
extendedDTUI("categorical_covariates_distributions_summary")
),
tabPanel(
"Individual values",
extendedDTUI("categorical_covariates_individual_table")
)
))
)
))),
tabItem(tabName = "params_vs_covs_tab",
fluidRow(
box(
width = 12,
title = "Settings",
fluidRow(
column(
width = 2,
radioButtons(
"p_c_values_type",
"Individual parameters and covariates values",
choices = c("First value only" = 0, "All values" = 1),
inline = TRUE
)
),
column(
width = 2,
selectInput(
"p_c_parameters_type_selection",
"Parameters type",
choices = c(
"Random effects (ETAs)" = "eta",
"Individual parameters (POSTHOCs)" = "individual"
)
)
),
column(
width = 8,
selectizeInput(
"p_c_selected_parameters",
"Parameters selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select parameters...',
onInitialize = I('function() { this.setValue(""); }')
)
),
conditionalPanel(
condition = "input.covariates_type_tab == 'continuous_tab'",
selectizeInput(
"p_c_selected_continuous_covariates",
"Continuous covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
conditionalPanel(
condition = "input.covariates_type_tab == 'categorical_tab'",
selectizeInput(
"p_c_selected_categorical_covariates",
"Categorical covariates selection",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select covariates...',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
div(
align = "center",
actionButton("parameters_vs_covariates_refresh", "Refresh")
)
)
)
),
tabBox(
title = "Covariates",
id = "covariates_type_tab",
width = 12,
tabPanel(
title = "Continuous",
value = "continuous_tab",
fluidRow(
column(
width = 6,
flowLayout(
selectInput(
"parameters_covariates_correlations_method",
"Correlation method",
choices = c("pearson",
"kendall",
"spearman")
)
),
extendedPlotUI(
"parameters_covariates_correlations_plot",
zoomable = FALSE,
verbose = TRUE,
title = "parameters-covariates-correlations",
export_width = 210,
export_height = 210,
height = 700,
click = "parameters_covariates_correlations_plot_click"
)
),
column(
width = 6,
DT::dataTableOutput("parameters_covariates_correlations_table")
)
),
h4("Selected correlation"),
fluidRow(
column(
width = 6,
flowLayout(
selectInput("p_c_correlations_smooth", "Smoothing method", choices = smoothing_methods),
selectizeInput(
"p_c_correlations_group",
"Highlight by",
choices = NULL,
multiple = FALSE,
options = list(
placeholder = 'Select categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
),
extendedPlotUI("p_c_correlation_plot",
title = "parameters-covariates-correlations", height = 700)
),
column(width = 6,
DT::dataTableOutput("selected_p_c_points_table"))
)
),
tabPanel(
title = "Categorical",
value = "categorical_tab",
p(
"Select parameters and covariates and click the \"Refresh\" button"
),
extendedPlotUI(
"parameters_cat_covs_plot",
title = "parameters-vs-categorical-covariates",
height = 700
)
)
)
)),
tabItem(
tabName = "quality_criteria",
fluidRow(
box(
width = 8,
title = "Settings",
flowLayout(
uiOutput("qc_pred_type"),
checkboxInput("qc_log_data", "Log-transformed data", value = FALSE),
selectInput(
"qc_alpha",
"Confidence interval for tests",
choices = c(
"90%" = 0.10,
"95%" = 0.05,
"99%" = 0.01
),
selected = 0.05
),
selectizeInput(
"qc_split_by",
"Split by",
choices = NULL,
multiple = TRUE,
options = list(
placeholder = 'Select compartment or categorical covariate',
onInitialize = I('function() { this.setValue(""); }')
)
)
)
),
box(
width = 4,
title = "Report",
radioButtons(
"qc_report_format",
"Document format",
c("PDF", "HTML", "Word"),
inline = TRUE
),
fluidRow(column(
6, downloadButton("qc_download_report", "Download report")
),
column(
6, actionButton("qc_save_report", "Save in run files")
))
),
tabBox(
width = 12,
tabPanel(title = "Standard QC",
fluidRow(column(9,
DT::dataTableOutput("qc_standard"),
h4("Bias (Mean Prediction Error)"),
DT::dataTableOutput("qc_bias"),
h4("Precision (Root Mean Square Error)"),
DT::dataTableOutput("qc_precision")),
column(3,
h4("Formulas"),
tags$ul(
tags$li("Maximal Error",
withMathJax('$$ME=max(|pred_i-obs_i|)$$')),
tags$li(
"Absolute Average Fold Error",
withMathJax(
'$$AAFE=10^{\\dfrac{1}{N}{\\sum_{i=1}^N{|log(\\dfrac{pred_i}{obs_i}) |}}}$$'
)
),
tags$li("Mean Prediction Error",
withMathJax('$$MPE=\\dfrac{1}{N}{\\sum_{i=1}^N{pred_i-obs_i}}$$')),
tags$li("Mean Prediction Error (relative to the mean)",
withMathJax('$$MPE(\\%)=\\dfrac{MPE}{\\overline{obs}}$$')),
tags$li("Root Mean Square Error",
withMathJax('$$RMSE=\\sqrt{\\dfrac{\\sum_{i=1}^N{(pred_i-obs_i)^2}}{N}}$$')),
tags$li("RMSE (relative to the mean)",
withMathJax('$$RMSE(\\%)=\\dfrac{RMSE}{\\overline{obs}}$$'))
)))),
tabPanel(title = "Student's t-Tests",
em("Observations vs Predictions; paired, two-sided"),
DT::dataTableOutput("qc_t_test_obs"),
em("Residuals vs zero; paired, two-sided"),
DT::dataTableOutput("qc_t_test_res")),
tabPanel(title = "Correlation test",
em("Observations vs Predictions"),
DT::dataTableOutput("qc_corr_test")),
tabPanel(title = "Linear regression",
DT::dataTableOutput("qc_lin_reg"),
uiOutput("run_qc_lin_reg_plot")),
tabPanel(title = "R code",
shinyAce::aceEditor("qc_r_code", mode = "r", height = "100px", readOnly = TRUE))
)
)
),
tabItem(tabName = "outliers",
fluidRow(
box(
width = 8,
title = "Settings",
flowLayout(
uiOutput("outliers_cmt"),
uiOutput("outliers_restype"),
checkboxInput("outliers_avoid_zero", "Exclude baseline", value = TRUE),
conditionalPanel(
condition = "input.outliers_test_tab != 'boxplot'",
selectInput(
"outliers_pvalue",
"p-value threshold",
choices = c("1%" = 0.01, "5%" = 0.05, "10%" = 0.1),
selected = 0.05
)
)
)
),
box(
width = 4,
title = "Report",
radioButtons(
"outliers_report_format",
"Document format",
c("PDF", "HTML", "Word"),
inline = TRUE
),
fluidRow(column(
6,
downloadButton("outliers_download_report", "Download report")
),
column(
6, actionButton("outliers_save_report", "Save in run files")
))
),
tabBox(
width = 12,
id = "outliers_test_tab",
title = "Procedures",
tabPanel(title = "Normality check",
fluidRow(
column(
width = 8,
h4("Q-Q (Normal) plot"),
div(align="center",
extendedPlotUI(
"outliers_qqplot",
title = "outliers-qqplot",
width = 500,
height = 500,
export_width = 210,
export_height = 210
)
)
),
column(
width = 4,
h4("Kolmogorov-Smirnov test"),
h6("Note: duplicate values are considered as one unique value"),
DT::dataTableOutput("outliers_ks_table"),
br(),
verbatimTextOutput("outliers_ks_decision")
)
),
fluidRow(column(
12,
h4("Selected residuals"),
DT::dataTableOutput("selected_outliers_residuals_table")
))),
tabPanel(
title = "Grubb's test",
h4("Outliers"),
extendedDTUI("outliers_grubbs_table"),
br(),
verbatimTextOutput("outliers_grubbs_message"),
h6(
"Note: Grubb's test makes the assumption that data is normally distributed."
)
),
tabPanel(
title = "Non-parametric",
value = "boxplot",
flowLayout(
numericInput(
"outliers_k_coefficient",
"k coefficient (default = 3)",
value = 3,
min = 0,
step = 0.1
)
),
fluidRow(
column(
width = 6,
h4("Outliers"),
extendedDTUI("outliers_boxplot_table"),
br(),
verbatimTextOutput("outliers_boxplot_message")
),
column(
width = 6,
h4("Box-and-whisker plot"),
extendedPlotUI(
"outliers_boxplot",
title = "outliers-boxplot",
height = 600,
export_width = 210,
export_height = 210
)
)
),
h4("Note"),
div(
"Observations are detected as outliers if they are outside of the following interval:",
withMathJax(
"$$\\mathopen{[}F_f - k \\cdot IQR\\,;F_t + k \\cdot IQR\\mathclose{]}$$"
),
"with Ff being the first quartile, Ft the third quartile, IQR the inter-quartile range (Ft-Ff) and k is a coefficient (length of the whiskers as multiple of IQR)"
)
)
)
)),
tabItem(tabName = "vpc",
fluidRow(
box(
width = 12,
title = "Visual Predictive Checks",
tags$div(
"Based on the run estimation results, a control stream is generated to perform simulations for Visual Predictive Checks."
),
br(),
tags$strong("The following must be performed to generate VPC:"),
tags$ol(
tags$li("Set the local paths to NONMEM and NMTRAN check executables"),
tags$li("Set number of simulations and seed"),
tags$li("Generate the control stream for simulation"),
tags$li(
"Run simulations locally using NONMEM (do not close the application until computation is finished)"
)
),
br(),
tags$div("The VPC plots and data are then computed by the ", tags$strong("vpc"),
" R package (docs: ", tags$a(href = "http://vpc.ronkeizer.com/", target = "_blank", "http://vpc.ronkeizer.com/"), ")")
,
fluidRow(column(
8,
h4("NONMEM paths"),
fluidRow(column(6, uiOutput("nm_exe")),
column(6, uiOutput("nmcheck_exe")))
),
column(
4,
h4("Simulations settings"),
fluidRow(column(
6,
numericInput(
"vpc_n_simulations",
"Number of simulations",
value = 100,
min = 1,
width = 175
)
),
column(6,numericInput("vpc_seed", "Simulations seed", value = 123456, width = 175)
))
)),
tags$div(
align = "center",
actionButton("generate_control_stream", "Generate control stream")
),
uiOutput("vpc_location")
),
tabBox(
width = 12,
title = "VPC results",
tabPanel("Visualization",
fluidRow(
column(3,
fluidRow(
column(
6,
uiOutput("vpc_idv"),
uiOutput("vpc_cmt"),
checkboxInput("vpc_correction", "Prediction-correction", value = FALSE),
sliderInput(
"vpc_pi",
"Prediction interval (PI) range",
value = c(0.05, 0.95),
min = 0,
max = 1,
step = 0.01
),
sliderInput(
"vpc_ci",
"Confidence interval (CI) range",
value = c(0.05, 0.95),
min = 0,
max = 1,
step = 0.01
),
selectInput("vpc_bin_mid", "Bin mid", choices = c("mean", "middle")),
selectInput(
"vpc_binning",
"Binning method",
choices = c(
"data (default)" = "data",
"time" = "time",
"auto",
"density",
"jenks",
"none",
"pretty"
),
selected = "data"
),
numericInput("vpc_n_bins", "Number of bins", value = NA)
),
column(
6,
tags$label("Plot:"),
checkboxInput("vpc_obs_dv", "Observations", value = TRUE),
checkboxInput("vpc_obs_ci", "Observations CI", value = TRUE),
checkboxInput("vpc_show_pi", "Prediction interval", value = FALSE),
checkboxInput("vpc_pi_as_area", "PI as area", value = FALSE),
checkboxInput("vpc_pi_ci", "CI around PI", value = TRUE),
checkboxInput("vpc_obs_median", "Observations median", value = TRUE),
checkboxInput("vpc_sim_median", "Simulations median", value = FALSE),
checkboxInput("vpc_sim_median_ci", "Simulations median CI", value = TRUE),
checkboxInput("vpc_smooth", "Smooth", value = TRUE),
numericInput("vpc_lloq", "Lower limit of quantification", value = NA),
selectInput(
"vpc_x_scale",
"X-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
),
selectInput(
"vpc_y_scale",
"Y-axis scale",
choices = c("Linear" = "linear", "Logarithmic" = "log")
),
selectInput(
"vpc_facet_scale",
"Facet scales",
choices = c(
"Free" = "free",
"Free X" = "free_x",
"Free Y" = "free_y",
"Fixed" = "fixed"
),
selected = "fixed"
)
)
)),
column(
9,
fluidRow(column(2, uiOutput("vpc_strat1")), column(10, uiOutput(
"vpc_strat1_levels"
))),
fluidRow(column(2, uiOutput("vpc_strat2")), column(10, uiOutput(
"vpc_strat2_levels"
))),
extendedPlotUI("vpc_plot", title = "VPC", height = 700)
)
)),
tabPanel("Data",
fluidRow(
column(12,
extendedDTUI("vpc_db_table", title = "VPC summary data"),
extendedDTUI("vpc_obs_table", title = "Observations"),
extendedDTUI("vpc_sim_table", title = "VPC Simulations"))
))
)
)),
tabItem(tabName = "comparison",
fluidRow(
box(
width = 12,
title = "Compare run to...",
fluidRow(
column(
2,
actionButton("select_runs_to_compare", "Select runs"),
actionButton("clear_comparison_table", "Clear selection")
),
column(
10,
flowLayout(
uiOutput("comparison_estimation_n"),
uiOutput("comparison_estimation_status")
),
uiOutput("comparison_estimation_method")
)
)
),
div(
id = "comparison-content",
tabBox(
width = 12,
title = "Comparison",
tabPanel(
width = 12,
"Summary table",
h4("Termination status"),
DT::dataTableOutput("comparison_status_summary"),
br(),
h4("Detailed comparison"),
fluidRow(column(
10,
checkboxGroupInput(
"comparison_details",
"Details",
choices = c(
"Info" = "INFO",
"Summary" = "SUMMARY",
"THETA" = "THETA",
"OMEGA" = "OMEGA",
"SIGMA" = "SIGMA",
"ETABAR" = "ETABAR",
"SHRINKAGE" = "SHRINKAGE",
"Filenames" = "FILES"
),
selected = c("SUMMARY", "THETA", "OMEGA"),
inline = TRUE
)
),
column(
2,
downloadButton("export_comparison_table", "Export")
)),
extendedDTUI("comparison_table")
),
tabPanel("Plots",
fluidRow(
tabBox(
width = 12,
tabPanel(
"Parameters",
fluidRow(
column(
2,
selectInput(
"comparison_distribution_plot",
label = "Plot type",
choices = c(
"Histogram" = "histogram",
"Density" = "density",
"Boxplot" = "boxplot"
)
)
),
column(
2,
radioButtons(
"comparison_plots_details",
"Parameters",
choices = c(
"THETA" = "THETA",
"OMEGA" = "OMEGA",
"SIGMA" = "SIGMA"
),
selected = "THETA",
inline = TRUE
)
),
column(8, uiOutput("comparison_plots_parameters"))
),
extendedPlotUI("comparison_param_plot", height = 700)
),
tabPanel("OFV",
extendedPlotUI("comparison_ofv_plot", height = 700))
))),
tabPanel(
"Statistics",
checkboxGroupInput(
"comparison_stats_table_detail",
"Details selection",
choices = c("THETA", "OMEGA", "SIGMA", "ETABAR", "SHRINKAGE"),
selected = c("THETA", "OMEGA", "SIGMA"),
inline = TRUE
),
checkboxInput("comparison_stats_table_rse", "RSE", value = FALSE),
extendedDTUI("comparison_stats_table")
)
)
)
)),
tabItem(tabName = "table_files",
fluidRow(
box(
width = 12,
title = "Output tables",
uiOutput("run_table_selection"),
checkboxInput("subject_first_row", "Subject first row only", value = FALSE),
extendedDTUI("run_table_output")
)
)),
tabItem(tabName = "control_stream_file",
fluidRow(
tabBox(
width = 12,
title = "Control stream",
tabPanel(title = "Raw",
verbatimTextOutput("control_stream_raw_content")),
tabPanel(title = "PMXploit control stream object",
verbatimTextOutput("control_stream_parsed_content"))
)
)),
tabItem(tabName = "report_file",
fluidRow(
box(
width = 12,
title = "Report file (NONMEM output)",
verbatimTextOutput("report_content")
)
))
)
)
)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.