shiny integration in qraLm

DESCRIPTION

@@ -17,7 +17,7 @@ Description: qraLm: A R package for quantitative risk assessment for Listeria mo
 URL: https://github.com/vcadavez/qraLm
 BugReports: https://github.com/vcadavez/qraLm/issues
 Depends: R (>= 3.5.0)
-Imports: Rdpack, Hmisc, extraDistr, mc2d, matrixStats, DT, dplyr,
+Imports: Rdpack, Hmisc, extraDistr, mc2d, matrixStats, DT, dplyr, shiny,
          ggplot2, plotly, Rcpp, odeintr, doseresponsemodels
 LinkingTo: Rcpp, odeintr, BH
 RdMacros: Rdpack

NAMESPACE

@@ -32,6 +32,7 @@ export(fvTesting)
 export(plotLotsECDF.qraLm)
 export(plotRisk.qraLm)
 export(printMC.qraLm)
+export(runExample)
 export(sfBrineORsaltCC)
 export(sfBriningCC)
 export(sfCharacteristics)
@@ -48,6 +49,7 @@ export(sfSmearingCC)
 export(sfSmoking)
 export(sfTesting)
 export(summaryLot.qraLm)
+export(summaryRisk.qraLm)
 export(summaryServings.qraLm)
 export(summaryUnits.qraLm)
 importFrom(DT,datatable)
@@ -62,7 +64,9 @@ importFrom(dplyr,"%>%")
 importFrom(extraDistr,dbbinom)
 importFrom(extraDistr,rtbinom)
 importFrom(extraDistr,rtnorm)
+importFrom(matrixStats,rowMeans2)
 importFrom(matrixStats,rowSums2)
+importFrom(matrixStats,rowWeightedMeans)
 importFrom(mc2d,rdirichlet)
 importFrom(mc2d,rmultinomial)
 importFrom(mc2d,rpert)

R/Lot2LotGen.R

@@ -21,6 +21,7 @@
 #' @param propVarInter proportion of the total variance (\eqn{C0\_sdLog10^2}) attributed
 #'   to between-lot variance. The remaining (`1-propVarInter`) is the proportion of within-lot variance.
 #' @param Poisson need to be defined
+#' @param ... Other options used to control [LotGen()]
 #'
 #' @return A list of four elements:
 #'    \describe{

R/LotGen.R

@@ -1,17 +1,23 @@
 #' @title Generation of contaminated lots
 #'
 #' @description
-#' The function [LotGen()] generates a matrix of contaminated lots from information on the parameters of a normal distribution (mean, sd)
+#' The function [LotGen()] generates a matrix of contaminated lots from information on the 
+#' parameters of a normal distribution (mean, sd)
 #' representing the microbial concentration in the contaminated units (log10 CFU/g).
-#' Every row of the matrix corresponds to a production lot, broken down in units or portions, which are the columns of the matrix.
-#' There is no assumption that lots can have different prevalence values, and therefore the microbial numbers of all units or portions produced in a lot
+#' Every row of the matrix corresponds to a production lot, broken down in units or portions,
+#'  which are the columns of the matrix.
+#' There is no assumption that lots can have different prevalence values, and therefore the 
+#' microbial numbers of all units or portions produced in a lot
 #' are sampled from a microbial concentration distribution.
 #' @param nLots number of lots sampled or size of the Monte Carlo simulation (scalar).
 #' @param sizeLot number of units or portions produced in a lot (scalar).
 #' @param unitSize (`g`) weight of single unit or portion from a lot (scalar).
 #' @param P prevalence of contaminated lots (scalar)
-#' @param C0MeanLog (log10 CFU/g) mean parameter of the normal distribution representing the variability in the microbial concentration of contaminated units (scalar or vector).
-#' @param C0SdLog (log10 CFU/g) standard deviation parameter of the normal distribution representing the variability in the microbial concentration of contaminated units (scalar or vector).
+#' @param C0MeanLog (log10 CFU/g) mean parameter of the normal distribution representing
+#'  the variability in the microbial concentration of contaminated units (scalar or vector).
+#' @param C0SdLog (log10 CFU/g) standard deviation parameter of the normal distribution
+#'  representing the variability in the microbial concentration of contaminated units (scalar or vector).
+#' @param ... Other options used to control [LotGen()]
 #' @return A list of two elements:
 #'     \describe{
 #'              \item{N}{(`CFU`) A matrix of size `nLots` lots by `sizeLot` units representing the microbial numbers in the units or portions from contaminated lots;}

R/caPartitioningCC.R

@@ -1,41 +1,51 @@
 #' @title Cross-contamination of cantaloupe during dicing and partitioning into packed units
 #' 
 #' @description
-#' The [caPartitioningCC()] function simulates the potential cross-contamination of cantaloupes, when in direct contact with the dicing machine
+#' The [caPartitioningCC()] function simulates the potential cross-contamination of cantaloupes, 
+#' when in direct contact with the dicing machine
 #' or knives, followed by the partitioning of all dices produced in one processing lot (sublot) into packed units.
 #' The cross-contamination algorithm accounts for four possible scenarios:
 #' \enumerate{
 #'    \item cross-contamination occurring in sublots already contaminated; 
 #'    \item contamination occurring in sublots that were not contaminated;
 #'    \item no cross-contamination occurring in sublots already contaminated; and
-#'    \item no cross-contamination occurring in sublots that were not contaminated. Probabilities of occurrence of every event are computed.
+#'    \item no cross-contamination occurring in sublots that were not contaminated. Probabilities 
+#'    of occurrence of every event are computed.
 #'      }
-#' The partitioning algorithm randomly distributes the total numbers of cells from a contaminated sublot of dices into packed units. The dispersion factor `b`, 
-#' which is a parameter of the beta distribution, indicates the extent of cell clustering in the bulk of dices in the sublot, and ultimately
+#' The partitioning algorithm randomly distributes the total numbers of cells from a contaminated sublot
+#'  of dices into packed units. The dispersion factor `b`, 
+#' which is a parameter of the beta distribution, indicates the extent of cell clustering in the bulk
+#'  of dices in the sublot, and ultimately
 #' the heterogeneity in the number of cells distributed among pack units.
 #'
 #' @param data a list of:
 #' \describe{
-#'    \item{`N`}{(`CFU`) A matrix of size `newMC` sublots by `sizeSublot` portions of dices from one cantaloupe representing the numbers of \emph{L. monocytogenes} 
-#'    in dices;}
+#'    \item{`N`}{(`CFU`) A matrix of size `newMC` sublots by `sizeSublot` portions of dices from one 
+#'    cantaloupe representing the numbers of \emph{L. monocytogenes} in dices;}
 #'    \item{`P`}{Prevalence of contaminated sublots (scalar).}
 #'          }
 #' @param probCCDice Probability of cross-contamination from the dicing machine (scalar).
-#' @param trDicerMean Mean of the transfer coefficient of \emph{L. monocytogenes} from dicing machine to cantaloupe flesh (scalar or vector).
-#' @param trDicerSd Standard deviation of the transfer coefficient of \emph{L. monocytogenes} from dicing machine to cantaloupe 
+#' @param trDicerMean Mean of the transfer coefficient of \emph{L. monocytogenes} from dicing
+#'  machine to cantaloupe flesh (scalar or vector).
+#' @param trDicerSd Standard deviation of the transfer coefficient of \emph{L. monocytogenes} 
+#' from dicing machine to cantaloupe 
 #'  flesh (scalar or vector).
-#' @param nDicer (`CFU`) Numbers of \emph{L. monocytogenes} on the surface of the dicing machine ready to be transferred (scalar or vector).
+#' @param nDicer (`CFU`) Numbers of \emph{L. monocytogenes} on the surface of the dicing machine 
+#' ready to be transferred (scalar or vector).
 # #' @param newMC Number of sublots (scalar).
+#' @param nLots See [Lot2LotGen()].
 #' @param sizeLot Number of cantaloupes from a harvested lot (scalar).
 #' @param sizeSublot Number of cantaloupes processed in a sublot. It should be a multiple of `sizeLot` (scalar).
 #' @param cantaWeight (`g`) weight of a cantaloupe
 #' @param pulpYield (`%`) Cantaloupe pulp yield (scalar or vector).
 #' @param unitSize (`g`) Weight of a pack of cantaloupe dices (scalar).
-#' @param b Dispersion factor of the beta distribution representing the degree of heterogeneity in the number of cells between pack units (scalar).
+#' @param b Dispersion factor of the beta distribution representing the degree of heterogeneity in the number
+#'  of cells between pack units (scalar).
 #'
 #' @return A list with four elements:
 #' \describe{
-#'       \item{`N`}{(`CFU`) A matrix of size `newMC` sublots by Number_packs number of packs containing the numbers of \emph{L. monocytogenes} in a pack,
+#'       \item{`N`}{(`CFU`) A matrix of size `newMC` sublots by Number_packs number of packs containing 
+#'       the numbers of \emph{L. monocytogenes} in a pack,
 #'            from contaminated lots;}
 #'       \item{`P`}{Prevalence of contaminated sublots (scalar);} 
 #'       \item{`sizeSublot`}{Number of cantaloupes processed in a sublot (scalar);} 
@@ -63,8 +73,9 @@
 #' @export
 #' 
 #' @note The value of \eqn{beta = 1} represents moderate clustering of cells in the bulk of diced cantaloupe from a 
-#' sublot \insertCite{Nauta2005;textual}{qraLm}. \insertCite{Hoelzer2012;textual}{qraLm} established the log 10 of the transfer coefficient
-#' of \emph{L. monocytogenes} from knives to vegetables as a normal distribution with \eqn{TRdicer\_mean = -1.42} and \eqn{TRdicer\_sd = 0.52}. 
+#' sublot \insertCite{Nauta2005;textual}{qraLm}. \insertCite{Hoelzer2012;textual}{qraLm} established the log 10 of 
+#' the transfer coefficient of \emph{L. monocytogenes} from knives to vegetables as a normal distribution with
+#' \eqn{TRdicer\_mean = -1.42} and \eqn{TRdicer\_sd = 0.52}. 
 #'
 #' @examples
 #' library(extraDistr)

R/plotLotsECDF.qraLm.R

@@ -2,7 +2,7 @@
 #'
 #' @title plotLotsECDF Generic plot function to plot the ECDF
 #' @param x qraLm object see [Lot2LotGen()]
-#' @param ... optional plot parameters passed to the plot function
+#' @param ... Optional plot parameters passed to the [plotLotsECDF()] function
 #' @author Vasco Cadavez
 #'
 #' @importFrom stats weighted.mean
@@ -12,25 +12,25 @@
 #' @examples
 #'
 #' prod <- Lot2LotGen(
-#'   nLots = 1000,
-#'   sizeLot = 1000,
-#'   unitSize = 500,
-#'   betaAlpha = 0.5112,
-#'   betaBeta = 9.959,
-#'   C0MeanLog = 1.023,
-#'   C0SdLog = 0.3267,
-#'   propVarInter = 0.7
-#' )
+#'                    nLots = 1000,
+#'                    sizeLot = 1000,
+#'                    unitSize = 500,
+#'                    betaAlpha = 0.5112,
+#'                    betaBeta = 9.959,
+#'                    C0MeanLog = 1.023,
+#'                    C0SdLog = 0.3267,
+#'                    propVarInter = 0.7
+#'                    )
 #' prod1 <- Lot2LotGen(
-#'   nLots = 1000,
-#'   sizeLot = 1000,
-#'   unitSize = 500,
-#'   betaAlpha = 0.9112,
-#'   betaBeta = 2.959,
-#'   C0MeanLog = 2.023,
-#'   C0SdLog = 0.5267,
-#'   propVarInter = 0.7
-#' )
+#'                     nLots = 1000,
+#'                     sizeLot = 1000,
+#'                     unitSize = 500,
+#'                     betaAlpha = 0.9112,
+#'                     betaBeta = 2.959,
+#'                     C0MeanLog = 2.023,
+#'                     C0SdLog = 0.5267,
+#'                     propVarInter = 0.7
+#'                     )
 #'
 #' plotLotsECDF.qraLm(prod1)
 #'

R/runExample.R

@@ -0,0 +1,22 @@
+#' @title Shiny application for `qraLm` package
+#' 
+#' @description
+#' Shiny application to simulate quantitative risk assessment for *L. monocytogenes* 
+#' in Frozen Vegetables, Diced RTE Cantaloupe and Cold-smoked RTE Fish.
+#'
+#' @author Vasco Cadavez \email{[email protected]}
+#'
+#' @references
+#' \insertRef{stats}{qraLm}
+#'
+#' @importFrom Rdpack reprompt
+#'
+#' @export
+runExample <- function() {
+  appDir <- system.file("shiny-examples", "qraLmShiny", package = "qraLm")
+  if (appDir == "") {
+    stop("Could not find example directory. Try re-installing `qraLm`.", call. = FALSE)
+  }
+
+  shiny::runApp(appDir, display.mode = "normal")
+}

R/summaryLot.qraLm.R

@@ -1,4 +1,4 @@
-#' Print summary MC results
+#' Print summary MC results at Lot level
 #'
 #' @title summaryLot Generic print function to print the between lots MC summary statistics
 #' @param x qraLm object. See [Lot2LotGen()]
@@ -12,15 +12,15 @@
 #'
 #' @examples
 #' prod <- Lot2LotGen(
-#'   nLots        = 1000,
-#'   sizeLot      = 1000,
-#'   unitSize     = 500,
-#'   betaAlpha    = 0.5112,
-#'   betaBeta     = 9.959,
-#'   C0MeanLog    = 1.023,
-#'   C0SdLog      = 0.3267,
-#'   propVarInter = 0.7,
-#'   Poisson      = FALSE
+#'                    nLots        = 1000,
+#'                    sizeLot      = 1000,
+#'                    unitSize     = 500,
+#'                    betaAlpha    = 0.5112,
+#'                    betaBeta     = 9.959,
+#'                    C0MeanLog    = 1.023,
+#'                    C0SdLog      = 0.3267,
+#'                    propVarInter = 0.7,
+#'                    Poisson      = FALSE
 #' )
 #'
 #' summaryLot.qraLm(prod)
@@ -33,62 +33,64 @@ summaryLot.qraLm <- function(x, ...) {
   if (exists("unitSize", x) == TRUE) {
     lotN <- rowMeans(x$N / x$unitSize, na.rm = TRUE)
   } else {
-    lotN <- rowMeans(x$N, na.rm = TRUE)
+     lotN <- rowMeans(x$N, na.rm = TRUE)
   }
 
   if (exists("ProbUnitPos", x) == TRUE) {
-    NStats <- Hmisc::wtd.quantile(lotN,
-      weights = x$ProbUnitPos,
-      probs = c(0.0, 0.5, 0.025, 0.975, 1.0),
-      normwt = TRUE, na.rm = TRUE
-    )
-    NStatsMean <- stats::weighted.mean(lotN, w = x$ProbUnitPos, na.rm = TRUE)
+      NStats <- Hmisc::wtd.quantile(lotN,
+                                    weights = x$ProbUnitPos,
+                                    probs = c(0.0, 0.5, 0.025, 0.975, 1.0),
+                                    normwt = TRUE, na.rm = TRUE)
+      
+      NStatsMean <- stats::weighted.mean(lotN, w = x$ProbUnitPos, na.rm = TRUE)
   } else {
     NStats <- Hmisc::wtd.quantile(lotN,
-      weights = rep(1, nrow(x$N)),
-      probs = c(0.00, 0.0, 0.025, 0.975, 1.00),
-      normwt = TRUE, na.rm = TRUE
-    )
+                                  weights = rep(1, nrow(x$N)),
+                                  probs = c(0.00, 0.0, 0.025, 0.975, 1.00),
+                                  normwt = TRUE, na.rm = TRUE)
+
     NStatsMean <- mean(lotN, na.rm = TRUE)
   }
 
-  NStatsMin <- NStats[1]
+  NStatsMin    <- NStats[1]
   NStatsMedian <- NStats[2]
-  Q2.5 <- NStats[3]
-  Q97.5 <- NStats[4]
-  NStatsMax <- NStats[5]
+  Q2.5         <- NStats[3]
+  Q97.5        <- NStats[4]
+  NStatsMax    <- NStats[5]
+
   Counts <- rbind(
-    unname(NStatsMin),
-    unname(Q2.5),
-    unname(NStatsMean),
-    unname(NStatsMedian),
-    unname(Q97.5),
-    unname(NStatsMax)
-  )
-  #  log10Counts <- round(log10(ceiling(Counts)), digits=4)
-  log10Counts <- round(log10(Counts), digits = 4)
+                  unname(NStatsMin),
+                  unname(Q2.5),
+                  unname(NStatsMean),
+                  unname(NStatsMedian),
+                  unname(Q97.5),
+                  unname(NStatsMax)
+                  )
+  # logs function
+  log_var <- function(x) {
+    ifelse(x != 0, log10(x), 0)
+  }
+
+   log10Counts <- round(log_var(Counts), digits = 6)
   Statistics <- c("Minimum", "pct 2.5th", "Mean", "Median", "pct 97.5th", "Maximum")
   MCstats <- data.frame(Statistics, Counts, log10Counts)
 
-
   if (exists("ProbUnitPos", x) == TRUE) {
     names(MCstats) <- c("Statistics", "CFU/g", "log10 CFU/g")
     DT::datatable(MCstats,
-      caption = "Summary statistics: between lots mean counts",
-      class = "display",
-      fillContainer = FALSE,
-      options = list(dom = "t")
-    ) |>
+                  caption = "Summary statistics: between lots mean counts",
+                  class = "display",
+                  fillContainer = FALSE,
+                  options = list(dom = "t")) |>
       DT::formatSignif(columns = c("CFU/g", "log10 CFU/g"), digits = 4)
   } else {
     names(MCstats) <- c("Statistics", "CFU/Melon", "log10 CFU/Melon")
 
     DT::datatable(MCstats,
-      caption = "Summary statistics: between lots mean counts",
-      class = "display",
-      fillContainer = FALSE,
-      options = list(dom = "t")
-    ) |>
+                  caption = "Summary statistics: between lots mean counts",
+                  class = "display",
+                  fillContainer = FALSE,
+                  options = list(dom = "t")) |>
       DT::formatSignif(columns = c("CFU/Melon", "log10 CFU/Melon"), digits = 4)
   }
 }