Hex stickr (#183)

* make ggplot of distributions

* good rough draft

* tweaks

* readme update

* light tweaks

* update wordlist

* [skip actions] Restyle files

* wip fixing the superlinter

* fix markdown

* new colours

* update wordlist

* [skip actions] Restyle files

Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: gravesti <isaac.gravestock@roche.com>
Co-authored-by: gravesti <83659704+gravesti@users.noreply.github.com>

Author: 3 people

.pre-commit-config.yaml

@@ -81,4 +81,4 @@ repos:
     rev: v0.32.2
     hooks:
       - id: markdownlint
-        args: ["--ignore=tests/testthat/_snaps/*"]
+        args: ["--ignore=tests/testthat/_snaps/*", "--disable=MD033"]

README.md

@@ -1,34 +1,71 @@
-# psborrow2
+# psborrow2 <img src="./inst/img/psborrow2_hex.png" align="right" width="120" />
+
+<!-- badges: start -->
+
+[![Version](https://img.shields.io/static/v1.svg?label=github.com/genentech&message=v.0.0.1&color=DC0073)](https://github.com/Genentech/psborrow2)
+[![Lifecycle:
+maturing](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://www.tidyverse.org/lifecycle/#experimental)
+
+<!-- badges: end -->
 
 ## Overview
 
-The `psborrow2` package exists to provide an efficient framework for users to
-conduct Bayesian Dynamic Borrowing (BDB) analyses (for more information on BDB,
-see Ibrahim et al., 2000; Hobbs et al., 2012; Schmidli et al., 2014;
-Lewis, et al., 2019).
+`psborrow2` provides an efficient framework for users to
+conduct Bayesian Dynamic Borrowing (BDB) analyses and other
+innovative analytical designs.[^1] [^2]
+`psborrow2` has two main objectives:
+
+1. **Facilitate BDB analyses**. `psborrow2` has a user-friendly interface for
+   conducting BDB analyses that handles the computationally-difficult MCMC sampling
+   on behalf of the user.
 
-`psborrow2` has two main goals: First, facilitate simulation studies that
-compare different borrowing parameters (e.g. full borrowing, no borrowing, BDB)
-and other trial and BDB characteristics (e.g. sample size, covariates)
-in a unified way; and second, provide a user-friendly interface for applying
-BDB on the study results that handles the MCMC sampling on behalf of the user.
+2. **Facilitate simulation studies of BDB**. `psborrow2` includes a
+   framework to compare different trial and BDB characteristics in a unified way
+   in simulation studies to inform trial design.
 
-`psborrow2` is the successor of
-[`psborrow`](https://github.com/Genentech/psborrow). Major updates to
-`psborrow2` include:
+## Background
 
-* New MCMC software (STAN)
-* New user interface
-* Additional functionality
+`psborrow2` is the successor to
+[`psborrow`](https://github.com/Genentech/psborrow). `psborrow` is still freely
+available on [`CRAN`](https://cran.r-project.org/package=psborrow) with the
+same validated functionality; however, the package is not actively developed.
+Major updates in `psborrow2` include:
+
+- New, more flexible user interface
+- New MCMC software (STAN)
+- Expanded functionality
+
+The name `psborrow` combines propensity scoring (`ps`) and Bayesian dynamic
+`borrow`ing. As the name implies, this package can be used to combine dynamic
+borrowing and propensity-score adjustment/weighting methods.
 
 ## Installation
 
-`psborrow2` is currently under development and should therefore be used with
-caution. You can install the development version via:
+You can install the latest version of `psborrow2` with:
 
 ```r
 remotes::install_github("Genentech/psborrow2")
 ```
 
-Please note that this package requires [`cmdstanr`](https://mc-stan.org/cmdstanr/)
-to be installed.
+Please note that this package requires [`cmdstanr`](https://mc-stan.org/cmdstanr/).
+
+## Vignettes
+
+To learn how to use the `psborrow2` R package, refer to the package vignettes:
+
+```r
+browseVignettes("psborrow2")
+```
+
+## Bibliography
+
+[^1]:
+    Lewis CJ, Sarkar S, Zhu J, Carlin BP. Borrowing from historical control data
+    in cancer drug development: a cautionary tale and practical guidelines.
+    Statistics in biopharmaceutical research. 2019 Jan 2;11(1):67-78.
+
+[^2]:
+    Viele K, Berry S, Neuenschwander B, Amzal B, Chen F, Enas N, Hobbs B,
+    Ibrahim JG, Kinnersley N, Lindborg S, Micallef S. Use of historical control
+    data for assessing treatment effects in clinical trials. Pharmaceutical
+    statistics. 2014 Jan;13(1):41-54.

dev/hex_sticker.R

@@ -0,0 +1,71 @@
+# Color palette reference: https://coolors.co/04e762-f5b700-00a1e4-dc0073-89fc00
+
+# Make a ggplot2 image of 3 distributions
+make_ggplot_of_dists <- function() {
+  # Dependancies
+  require(ggplot2)
+  require(showtext)
+  require(dplyr)
+  require(ggthemes)
+
+  # Three normal distributions
+  log_x_vals <- seq(-15, 15, .01)
+  x_vals <- exp(log_x_vals)
+  norm_1 <- norm_2 <- norm_3 <- vector("numeric", length(log_x_vals))
+  norm_1 <- dnorm(x = log_x_vals, mean = 3, sd = 3)
+  norm_2 <- dnorm(x = log_x_vals, mean = .25, sd = 1.5)
+  norm_3 <- dnorm(x = log_x_vals, mean = 2, sd = .9)
+
+  # Combine
+  df <- data.frame(
+    density = c(norm_1, norm_2, norm_3),
+    x = rep(log_x_vals, 3),
+    analysis = factor(rep(1:3, each = length(x_vals)))
+  )
+
+  # Plot
+  ggplot(df) +
+    geom_density(
+      aes(
+        x = x,
+        y = density,
+        fill = analysis,
+        color = analysis
+      ),
+      stat = "identity",
+      alpha = 0.8,
+      show.legend = FALSE
+    ) +
+    scale_fill_manual(values = c("#F5B700", "#00A1E4", "#DC0073")) +
+    scale_color_manual(values = c("#F5B700", "#00A1E4", "#DC0073")) +
+    theme_void()
+}
+
+make_hexplot <- function(out_path = "./inst/img/psborrow2_hex.png") {
+  # Dependancies
+  require(hexSticker)
+  require(showtext)
+
+  # HexSticker
+  hexSticker::sticker(
+    subplot = make_ggplot_of_dists(),
+    package = "psborrow2",
+    p_size = 140,
+    p_color = "#094F26",
+    p_y = 1.4,
+    s_y = .83,
+    s_x = .7,
+    s_width = 4.5,
+    s_height = .7,
+    h_fill = "#b7fed4",
+    h_color = "#094F26",
+    h_size = 2,
+    url = "github.com/Genentech/psborrow2",
+    u_size = 24,
+    filename = out_path,
+    p_family = "mono",
+    dpi = 2000
+  )
+}
+
+make_hexplot()

inst/WORDLIST

@@ -2,6 +2,7 @@ ATT
 Acknowledgements
 Amzal
 BDB
+CJ
 BayesPPD
 Bove
 CMD
@@ -13,14 +14,17 @@ IPTW
 Ibrahim
 Jiawen
 Kinnersley
+Lifecycle
 Lindborg
 MatchIt
 Micallef
 Neuenschwander
+PSweight
 Pharmaceut
 RCT
 Roychoudhury
 Sabanes
+Sarkar
 Schmidli
 Statist
 Viele
@@ -32,6 +36,7 @@ approxeq
 bernoulli
 binarization
 binarizing
+biopharmaceutical
 cauchy
 cmdstanr
 commensurability
@@ -51,6 +56,7 @@ https
 hyperprior
 hyperpriors
 infty
+ing
 interpretable
 mc
 operatorname