shiny-vpc

Shiny App for Visual Predictive Check (VPC)

This repo contains a Shiny application to derive VPCs and pcVPCs using traditional binning methods and advanced regression techniques (binless methods). The application utilizes the tidyvpc package developed by our team of pharmacometricians and software engineers at Certara.

Features

• Binning Methods: Binning methods include x-variable, pam, kmeans, quantiles, centers, breaks, and more.
• Binless Methods(Advanced Regression Techniques): Additive Quantile Regression to generate VPC and LOESS pcVPC.
• Binning/Binless by Strata: Specify binning and binless parameters for each level of stratification.
• Plot Customization: Use ggplot2 to specify plot facetting, color aesthetics, line types, and more.
• Code Generation: Generate vpcstats and ggplot2 code to reproduce VPC in your local R Session.

Installation and Running information

# Install the development version of tidyvpc from GitHub:
install.packages("devtools")
library(devtools)
devtools::install_github("jameswcraig/tidyvpc")
# Install CRAN Dependencies:
install.packages(c("backports", "DT", "ggplot2", "rlang", "shinyAce",
  "shinydashboard", "shinydashboardPlus", "shinyjs",
  "shinycssloaders", "shinyWidgets"))
# Install shinyMeta by Rstudio for code generation:
library(remotes)
remotes::install_github("rstudio/shinymeta")
# Run Shiny App for Visual Predictive Check (VPC)
shiny::runGitHub("shiny-vpc", "jameswcraig")

File Layout

The Shiny application requires 2 separate file uploads which can be in csv, fit, or RDS format:

1.) obs_data: Observed data with the following columns:

• Required: TIME, DV
• Optional: PRED, LLOQ, Stratification Variables

2.) sim_data: Simulated data with the following columns:

• Required: TIME, DV, REP
• Optional: PRED, LLOQ, Stratification Variables

Notes: Please subset MDV = 0 in data before uploading and make sure there are no missing rows. Variable names do not need to match the names above but at a minimum your data will need to contain a y-variable (typically named DV) and x-variable (typically TIME) to derive VPC.

Example Code Generation

library(vpcstats)
library(ggplot2)

vpc <- observed(obs_data, x = !!rlang::sym("TIME"), y = !!rlang::sym("DV")) %>%
  simulated(sim_data, y = !!rlang::sym("DV")) %>%
  censoring(blq = !!rlang::sym("DV") < !!rlang::sym("LLOQ"), lloq = !!rlang::sym("LLOQ")) %>%
  stratify(~ISM) %>%
  binless(qpred = c(0.1, 0.5, 0.9), optimize = TRUE, optimization.interval = c(0L, 7L), conf.level = 0.95) %>%
  vpcstats()

vpcPlot <- ggplot(vpc$stats, aes(x = x)) + 
	facet_grid(~ISM, scales = "free", as.table = FALSE) + 
	geom_ribbon(aes(ymin = lo, ymax = hi, fill = qname, col = qname, group = qname), alpha = 0.1, col = NA) + 
	geom_line(aes(y = md, col = qname, group = qname)) + geom_line(aes(y = y, linetype = qname), size = 1) + 
	geom_hline(data = unique(vpc$data[, .(LLOQ), by = eval("ISM")]), aes(yintercept = !!as.symbol("LLOQ")), linetype = "dotted", size = 1) + 
	geom_text(data = unique(vpc$data[, .(LLOQ), by = eval("ISM")]), aes(x = 10, y = LLOQ, label = paste("LLOQ", LLOQ, sep = "="), ), vjust = -1) + 
	scale_colour_manual(name = "Simulated Percentiles\nMedian (lines) 95% CI (areas)", breaks = c("q0.1", "q0.5", "q0.9"), values = c("red", "blue", "red"), labels = c("10%", "50%", "90%")) + 
	scale_fill_manual(name = "Simulated Percentiles\nMedian (lines) 95% CI (areas)", breaks = c("q0.1", "q0.5", "q0.9"), values = c("red", "blue", "red"), labels = c("10%", "50%", "90%")) + 
	scale_linetype_manual(name = "Observed Percentiles\nMedian (lines) 95% CI (areas)", breaks = c("q0.1", "q0.5", "q0.9"), values = c("dotted", "solid", "dashed"), labels = c("10%", "50%", "90%")) + 
	guides(fill = guide_legend(order = 2), colour = guide_legend(order = 2), linetype = guide_legend(order = 1)) + 
	theme(legend.position = "top", legend.key.width = grid::unit(2, "cm")) + 
	labs(x = "TIME", y = "Concentration")