move data-raw/helpers.R into package

clean up license
update targets pipelines to depend on package

.Rbuildignore

@@ -17,7 +17,7 @@ coverage.*
 init.sh
 workflows.md
 images
-^\_targets\.R$
-^\_targets$
+^\_targets\.yaml$
 ^README\.Rmd$
 ^data-raw$
+^LICENSE\.md$

DESCRIPTION

@@ -5,11 +5,11 @@ Version: 0.0.1
 Date: 2023-11-09
 Authors@R: person("Jonathan","Sidi" , , "[email protected]", role = c("aut", "cre"))
 Description: Reproducible simulated clinical trials for use in R packages.
-License: Apache License 2.0 | file LICENSE
+License: MIT + file LICENSE
 URL: https://github.com/openpharma/clindata
 BugReports: https://github.com/openpharma/clindata/issues
 Depends: R (>= 4.0)
-Imports: tools, utils
+Imports: tools, utils, MASS, clusterGeneration
 Suggests: 
     knitr,
     withr,

LICENSE

@@ -1,13 +1,2 @@
-Copyright 2022 F. Hoffmann-La Roche AG
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
+YEAR: 2024
+COPYRIGHT HOLDER: clindata authors

LICENSE.md

@@ -0,0 +1,21 @@
+# MIT License
+
+Copyright (c) 2024 clindata authors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

NAMESPACE

@@ -6,6 +6,10 @@ export(cache_dir)
 export(cache_init)
 export(cache_ls)
 export(cache_rm)
+export(compute_unstructured_matrix)
+export(generate_outcomes)
 export(list_data)
 export(load_data)
+export(mar)
+export(mcar)
 export(read_data)

data-raw/pipelines/helpers.R → R/simulations.R

@@ -1,8 +1,5 @@
-#' @importFrom stats cov2cor plogis rbinom
-#' @importFrom clusterGeneration genPositiveDefMat
-#' @importFrom MASS mvrnorm
-
 # Helper function for randomly generating unstructured covariance matrix.
+#' @export
 compute_unstructured_matrix <- function(
     vars = seq(from = 1, by = 0.5, length.out = 10)) {
   n_visits <- length(vars)
@@ -14,19 +11,32 @@ compute_unstructured_matrix <- function(
   return(us_mat)
 }
 
-pad_number <- function(x, width = 2, pad = "0") {
-  fmt <- sprintf("%%%s%ss", pad, width)
-  sprintf(fmt, x)
-}
-
 # MCAR helper function.
+#' @export
 mcar <- function(data, col, p) {
   missing_vec <- stats::rbinom(nrow(data), 1, p)
   data[missing_vec == 1, col] <- NA_real_
   data
 }
 
+# Helper function for generating data.
+#' @export
+generate_outcomes <- function(model_mat, cov_mat, effect_coefs, effect_var) {
+
+  # Generate the outcomes.
+  n_visits <- nrow(cov_mat)
+  n_obs <- nrow(model_mat) / n_visits
+
+  rbetas <- MASS::mvrnorm(1, effect_coefs, diag(effect_var))
+  mu <- as.numeric(model_mat %*% rbetas)
+
+  resid <- as.numeric(MASS::mvrnorm(n_obs, rep(0, n_visits), cov_mat))
+  y_hat <- mu + resid
+  as.vector(y_hat)
+}
+
 # MAR helper function.
+#' @export
 mar <- function(data, type, fn = NULL) {
   # Compute missingness probabilities.
 
@@ -51,18 +61,3 @@ mar <- function(data, type, fn = NULL) {
   # Only keep non-missing visits.
   data[missing_ind == 0, ]
 }
-
-# Helper function for generating data.
-generate_outcomes <- function(model_mat, cov_mat, effect_coefs, effect_var) {
-
-  # Generate the outcomes.
-  n_visits <- nrow(cov_mat)
-  n_obs <- nrow(model_mat) / n_visits
-
-  rbetas <- MASS::mvrnorm(1, effect_coefs, diag(effect_var))
-  mu <- as.numeric(model_mat %*% rbetas)
-
-  resid <- as.numeric(MASS::mvrnorm(n_obs, rep(0, n_visits), cov_mat))
-  y_hat <- mu + resid
-  as.vector(y_hat)
-}

data-raw/_targets.R

@@ -4,7 +4,7 @@ library(targets)
 # Set target options:
 tar_option_set(
   # packages that your targets need to run
-  packages = c("usethis", "tibble", "dplyr", "stats", "clusterGeneration", "MASS")
+  packages = c("usethis", "tibble", "dplyr", "stats", "clindata")
 )
 
 pipelines_path <- here::here("data-raw/pipelines")

data-raw/pipelines/bcva/bcva.R

@@ -1,5 +1,4 @@
 source(here::here("data-raw/pipelines/bcva/bcva-helpers.R"))
-source(here::here("data-raw/pipelines/helpers.R"))
 
 # BCVA data-generating process.
 
@@ -15,7 +14,7 @@ pipe_bcva <- list(
     )
   }),
   # Generate Covariate Matrix
-  tar_target(bcva_outcome_covar_mat, compute_unstructured_matrix()),
+  tar_target(bcva_outcome_covar_mat, clindata::compute_unstructured_matrix()),
   # Generate the covariates.
   tar_target(bcva_covars_tbl, {
     bcva_generate_covariates(
@@ -25,7 +24,7 @@ pipe_bcva <- list(
   }),
   # Generate the outcomes.
   tar_target(bcva_outcomes, {
-    generate_outcomes(
+    clindata::generate_outcomes(
       model_mat = bcva_model_mat(bcva_covars_tbl),
       cov_mat = bcva_outcome_covar_mat,
       effect_coefs = bcva_coefs(
@@ -48,18 +47,18 @@ pipe_bcva <- list(
   }),
   # Delete observations at random.
   tar_target(bcva_tbl_missing, {
-    mar(bcva_tbl, type = bcva_scenario$missing_type, bcva_lin_pred)
+    clindata::mar(bcva_tbl, type = bcva_scenario$missing_type, bcva_lin_pred)
   }),
   # Format to resemble BCVA dataset.
   tar_target(bcva_data, {
     bcva_tbl_missing |>
       dplyr::transmute(
         USUBJID = factor(participant),
         VISITN = visit_num,
-        AVISIT = paste0("VIS", pad_number(visit_num)),
+        AVISIT = paste0("VIS", sprintf("%02s", visit_num)),
         AVISIT = factor(
           AVISIT,
-          levels = paste0("VIS", pad_number(seq_len(10)))
+          levels = paste0("VIS", sprintf("%02s", seq_len(10)))
         ),
         ARMCD = ifelse(trt == 1, "TRT", "CTL"),
         RACE = ifelse(strata == 1, "Black",

data-raw/pipelines/fev/fev.R

@@ -1,4 +1,3 @@
-source(here::here('data-raw/pipelines/helpers.R'))
 source(here::here('data-raw/pipelines/fev/fev-helpers.R'))
 
 # FEV1 data-generating process.
@@ -16,7 +15,7 @@ pipe_fev <- list(
   tar_target(fev_outcome_covar_mat, {
     v11 <- rnorm(1, 40, 0.1)
     vars <- v11 * c(1, 2/3, 1/3, 5/2)
-    compute_unstructured_matrix(vars)
+    clindata::compute_unstructured_matrix(vars)
   }),
   # Generate the covariates.
   tar_target(fev_covars_tbl, {
@@ -27,7 +26,7 @@ pipe_fev <- list(
   }),
   # Generate the outcomes.
   tar_target(fev_outcomes, {
-    generate_outcomes(
+    clindata::generate_outcomes(
       model_mat = fev_model_mat(fev_covars_tbl),
       cov_mat = fev_outcome_covar_mat,
       effect_coefs = fev_coefs(trt_coef = fev_scenario$trt_coef),
@@ -43,7 +42,7 @@ pipe_fev <- list(
   }),
   # Delete observations at random.
   tar_target(fev_data, {
-    mcar(fev_tbl, "FEV1", fev_scenario$missing_percent)
+    clindata::mcar(fev_tbl, "FEV1", fev_scenario$missing_percent)
   }),
   tar_target(fev_deploy, {
     usethis::use_data(fev_data, overwrite = TRUE)