inst/shinyapp/ui.R

In andrewhooker/PopED: Population (and Individual) Optimal Experimental Design

library(shiny)
library(PopED)
library(rhandsontable)

ui <- function(){
  offset <- 0
  navbarPage(
    title="PopED - Population Experimental Design",
    collapsible = TRUE,
    tabPanel("Model Definition", fluidPage(
      fluidRow(
        column(
          9, offset = offset, 
          #8, offset = 2, #align ="center",
          #title = "Model Definition",
          #titlePanel("Model Definition"),
          h1("Structural Model"),
          helpText("Define the structural pharmacokinetic (PK) and/or the pharmacodynamic (PD) model.")
        )
      ),
      fluidRow(
        column(
          3, offset = offset, 
          checkboxInput("pk_mod", label = "PK model", value = TRUE),  
          #h4("PK Model"),
          conditionalPanel(
            condition = "input.pk_mod == true",
            selectInput("struct_PK_model", "Structural PK model:",
                      choices = list(
                        #"None"="NULL",
                        "1-compartment" = "ff.PK.1.comp.iv",
                        "1-compartment, 1st-order absorption" = "ff.PK.1.comp.oral",
                        "2-compartment" = "ff.PK.2.comp.iv",
                        "2-compartment, 1st-order absorption" = "ff.PK.2.comp.oral",
                        "3-compartment" = "ff.PK.3.comp.iv",
                        "3-compartment, 1st-order absorption" = "ff.PK.3.comp.oral"
                      ),
                      selected = "ff.PK.1.comp.oral"),
            selectInput("param_PK_model", "PK model parameterization:",
                        choices = list(
                          "Clearace and Volume"="CL",
                          "Rate constants" = "KE"
                        ),
                        selected = "cl")
          )
        ),
        #br(),
        
        column(3, offset = 0, 
               #h4("PD Model"),
               checkboxInput("pd_mod", label = "PD model", value = FALSE),
               conditionalPanel(
                 condition = "input.pd_mod == true",
                 selectInput("struct_PD_model", "Structural PD Model:",
                             list(
                               #"None" = "NULL",
                               "Linear" = "linear",
                               "Emax" = "emax",
                               "Emax with hill coefficient" = "hill"
                             ))
               ),
               conditionalPanel(
                 condition = "input.pd_mod == true && input.pk_mod == true",
                 selectInput("link_fcn", label = "Link Function", 
                             list(
                               "Direct Eeffect" = "direct",
                               "Effect compartment" = "effect",
                               "Turnover inhibit Kin"="inhib_kin",
                               "Turnover stimulate Kin"="stim_kin",
                               "Turnover inhibit Kout"="inhib_kout",
                               "Turnover stimulate Kout"="stim_kout"))
               )
        )
      ),
      fluidRow(
        column(9, offset = offset, 
               h1("Between Subject Variability Model"),
               helpText("Define the between subject variability (BSV) for the PK and/or PD model.")
        )
      ),
      fluidRow(
        column(3, offset = offset, 
               rHandsontableOutput("hot2")
        )
      ),
      
      #uiOutput("parameter_vales"),
      fluidRow(
        column(9, offset = offset, 
               h1("Residual Unexplained Variability Model"),
               helpText("Define the residual unexplained variability (RUV) for the PK and/or PD model.")
        )
      ),
      conditionalPanel(
        condition = "input.struct_PK_model != 'NULL'",
        ## TODO: add more structural models, create function like in PKPDmodels or with PKPDsim
        ## allow for different parameterizations as in PKPDsim (list to change values in equations)
        ## print out equation so that users can sse what they are using
        
        selectInput("ruv_pk_model", "PK RUV model",
                    list(
                      "Additive + Proportional" = "feps.add.prop",
                      "Proportional" = "feps.prop",
                      "Additive" = "feps.add"
                    ))
      ),
      conditionalPanel(
        condition = "input.pd_mod == true",        
        ## TODO: add more structural models, create function like in PKPDmodels or with PKPDsim
        ## allow for different parameterizations as in PKPDsim (list to change values in equations)
        ## print out equation so that users can sse what they are using
        
        selectInput("ruv_pd_model", "PD RUV model",
                    list(
                      "Additive + Proportional" = "feps.add.prop",
                      "Proportional" = "feps.prop",
                      "Additive" = "feps.add"
                    ))
      ),
      #hr(),
      
      
      #fluidRow(box(rHandsontableOutput("hot2", height = 400))),
      
      # h2("Between Subject Variability Model"),
      # conditionalPanel(
      #   condition = "input.pk_mod == true",
      #   
      #   selectInput("bsv_pk_model","PK BSV",
      #               list(
      #                 "Exponential" = "exp",
      #                 "Proportional" = "prop",
      #                 "Additive" = "add",
      #                 "None" = "none"
      #               )),
      #   checkboxInput("per_pd_param", label = "Choose per PK parameter", value = FALSE)  
      # ),
      # conditionalPanel(
      #   condition = "input.pd_mod == true",
      #   
      #   selectInput("bsv_pd_model","PD BSV",
      #               list(
      #                 "Exponential" = "exp",
      #                 "Proportional" = "prop",
      #                 "Additive" = "add",
      #                 "None" = "none"
      #               )),
      #   checkboxInput("per_pd_param", label = "Choose per PD parameter", value = FALSE)  
      # ),
      
      br()
      
      
    )),
    
    tabPanel("Parameter Definition",
             #uiOutput("parameter_vales"),
             #radioButtons("useType", "Use Data Types", c("TRUE", "FALSE")),
             
             fluidRow(
               column(9, offset = offset, 
                      h1("Parameter definition"),
                      helpText("Define the parameter values for the PK and/or PD model.")
               )
             ),
             
             h5("Fixed effects"),
             rHandsontableOutput("hot3"),
             hr(),
             h5("BSV parameters (variance units)"),
             rHandsontableOutput("hot4"),
             h5("Fixed BSV parameters"),
             rHandsontableOutput("hot5"),
             hr(),
             
             h5("RUV parameters (variance units)"),
             rHandsontableOutput("hot6"),
             h5("Fixed RUV parameters"),
             rHandsontableOutput("hot7"),
             
             rHandsontableOutput("hot"),
             #helpText(paste0("value is ")),
             #h3("Residual Unexplained Variability Model"),
             # conditionalPanel(
             #   condition = "input.bsv_per_param == false",
             #   selectInput("bsv_model","",
             #               list(
             #                 "Exponential" = "exp",
             #                 "Proportional" = "prop",
             #                 "Additive" = "add",
             #                 "None" = "none"
             #               ))
             # ),
             #checkboxInput("bsv_per_param", label = "Choose BSV model per parameter", value = FALSE),
             br()
    ),
    tabPanel("Design Definition",
             #textInput("num_groups", "Nunber of design groups", "1"),
             #uiOutput("group_designs"),
             #uiOutput("test"),
             h5("Dosing"),
             rHandsontableOutput("hot_dose_table"),
             h5("Sampling"),
             rHandsontableOutput("hot_sample_table"),
             #tableOutput('table_tmp'),
             hr(),
             #actionButton("update_design","Update design"),
             ##  create plot of model
            # plot_model_prediction(poped.db,IPRED=T,DV=T),

             #     checkboxInput("IPRED", "Show IPRED", FALSE),
             #
             #     checkboxInput("DV", "Show DV", FALSE),
             #     checkboxInput("separate.groups", "Separate Groups", FALSE)

             ## identifiers
             br()
    ),
    
    #mainPanel(
    #h3(textOutput("caption")),
    #     checkboxInput("smooth", "Smooth"),
    #     conditionalPanel(
    #       condition = "input.smooth == true",
    #       selectInput("smoothMethod", "Method",
    #                   list("lm", "glm", "gam", "loess", "rlm"))
    #     ),
    
    tabPanel("Plot model/design", 
             actionButton("update_plot","Create plot"),
             plotOutput("modelPlot"),
             submitButton("Update View"),
             checkboxInput("IPRED", "Show IPRED", FALSE),
             checkboxInput("DV", "Show DV", FALSE),
             checkboxInput("separate.groups", "Separate Groups", FALSE),
             br()
    ), 
    tabPanel("Evaluate design", verbatimTextOutput("summary"),
             br()
    ),
    tabPanel("Optimize design", tableOutput("table"),
             br()
    ),
    footer = list(
      fluidRow(
        column(12, offset = offset, align="center",
               hr(),
               img(src = "poped_splash.png", height = 72, width = 72),
               a(paste("PopED for R (", packageVersion("PopED"),")",sep=""), 
                 href = "http://poped.sf.net"),
               div(tags$small(
                 img(src = "", height = 2, width = 3),
                 "(c) 2014-2016, Andrew C. Hooker,",
                 tags$br(),
                 img(src = "", height = 2, width = 3),
                 "Pharmacometrics Research Group,",
                 tags$br(),
                 img(src = "", height = 2, width = 3),
                 "Uppsala University,",
                 tags$br(),
                 img(src = "", height = 2, width = 3),
                 "Sweden"
               ))
        )
      )
    )
  )
}

#plotOutput("modelPlot")
#   footer=(
#     br(),
#     img(src = "poped_splash.png", height = 72, width = 72), 
#     a(paste("PopED for R (", packageVersion("PopED"),")",sep=""), 
#       href = "http://poped.sf.net"),
#     h6("(c) 2014, Andrew C. Hooker, Pharmacometrics Research Group, Uppsala University, Sweden")
#     
#   )