Replace usage of scda and scda.2022 with random.cdisc.data (#64)

Closes #60

DESCRIPTION

@@ -27,9 +27,8 @@ URL: https://insightsengineering.github.io/biomarker-catalog/
 BugReports: https://github.com/insightsengineering/biomarker-catalog/issues
 Imports:
     pkgdown,
-    scda,
-    scda.2022,
-    teal.modules.clinical (>= 0.9.0.9014),
+    random.cdisc.data (>= 0.3.15),
+    teal.modules.clinical (>= 0.9.1),
     tern (>= 0.9.4),
     tern.mmrm,
     teal (>= 0.15.2),

graphs/DG1/setup.qmd

@@ -4,7 +4,7 @@ exclude-listing: true
 
 ## Setup
 
-We will use the `synthetic_cdisc_data$adsl` data set from the `scda` package and `ggplot2` to create the plots.
+We will use the `cadsl` data set from the `random.cdisc.data` package and `ggplot2` to create the plots.
 In this example, we will plot histograms of one or multiple numeric variables.
 We start by selecting the biomarker evaluable population with the flag variable `BEP01FL` and then populating a new continuous biomarker variable, `BMRKR3`.
 
@@ -15,7 +15,7 @@ library(dplyr)
 library(tibble)
 library(tidyr)
 
-adsl <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adsl %>%
+adsl <- random.cdisc.data::cadsl %>%
   df_explicit_na() %>%
   filter(BEP01FL == "Y") %>%
   mutate(BMRKR3 = rnorm(n(), mean = 7, sd = 2))

graphs/DG3/setup.qmd

@@ -6,7 +6,7 @@ exclude-listing: true
 
 The graphs below summarize the distribution of a categorical biomarker variable as barplots, either in the overall population or by one or more categorical clinical variables.
 
-We will use the `synthetic_cdisc_data$adsl` data set from the `scda` package to illustrate the graph and will select on the biomarker evaluable population with `BEP01FL`.
+We will use the `cadsl` data set from the `random.cdisc.data` package to illustrate the graph and will select on the biomarker evaluable population with `BEP01FL`.
 The column `BMRKR2` contains the biomarker values on a categorical scale.
 We will use `ARM` as categorical clinical variable.
 
@@ -15,7 +15,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adsl <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adsl %>%
+adsl <- random.cdisc.data::cadsl %>%
   df_explicit_na() %>%
   filter(BEP01FL == "Y")
 ```

graphs/KG1/kg01b.qmd

@@ -14,7 +14,7 @@ categories: [KG]
 Here we only filter the time-to-event dataset for the overall survival observations, but keep all treatment arms and the overall population.
 
 ```{r}
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(PARAMCD == "OS") %>%
   mutate(is_event = CNSR == 0)

graphs/KG1/setup.qmd

@@ -4,7 +4,7 @@ exclude-listing: true
 
 ## Setup
 
-We will use the `synthetic_cdisc_data$adtte` data set from the `scda` package to create the Kaplan-Meier (KM) plots.
+We will use the `cadtte` data set from the `random.cdisc.data` package to create the Kaplan-Meier (KM) plots.
 We start by filtering the time-to-event dataset for the overall survival observations and by one treatment arm (A), creating a new variable for event information, and curating a list of variables required to produce the plot.
 
 ```{r, message = FALSE}
@@ -13,7 +13,7 @@ library(dplyr)
 library(ggplot2)
 library(grid)
 
-adtte_arm <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte_arm <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(PARAMCD == "OS", ARM == "A: Drug X") %>%
   mutate(is_event = CNSR == 0)

graphs/KG2/setup.qmd

@@ -10,7 +10,7 @@ The same setup as in [KG1](../../graphs/KG1/kg01.qmd) is used.
 library(tern)
 library(dplyr)
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(PARAMCD == "OS") %>%
   mutate(is_event = CNSR == 0)

graphs/KG4/setup.qmd

@@ -11,7 +11,7 @@ Since we are using biomarker information, we filter on the biomarker evaluable p
 library(tern)
 library(dplyr)
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(PARAMCD == "OS" & BEP01FL == "Y") %>%
   mutate(

graphs/KG5/setup.qmd

@@ -11,7 +11,7 @@ The difference here is that we create the initial binary biomarker variable `BMR
 library(tern)
 library(dplyr)
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(PARAMCD == "OS" & BEP01FL == "Y") %>%
   mutate(

graphs/RFG1/setup.qmd

@@ -7,7 +7,7 @@ exclude-listing: true
 These templates are helpful when we are interested in the odds ratios between two groups, usually two treatment arms.
 We would like to assess how the odds ratio changes when we look at different subgroups, often defined by categorical biomarker variables, e.g.
 
-We will use the `synthetic_cdisc_data$adrs` data set from the `scda` package to create the response forest graph.
+We will use the `cadrs` data set from the `random.cdisc.data` package to create the response forest graph.
 We start by filtering the `adrs` data set for the Best Confirmed Overall Response by Investigator and patients with measurable disease at baseline `(BMEASIFL == "Y")`.
 We create a new variable for response information (we define response patients to include CR and PR patients), and binarize the `ARM` variable.
 We also fix a data artifact by setting the categorical biomarker variable `BMRKR2` to an explicit `<Missing>` level for the non-biomarker evaluable population.
@@ -19,7 +19,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adrs <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adrs %>%
+adrs <- random.cdisc.data::cadrs %>%
   df_explicit_na() %>%
   filter(PARAMCD == "BESRSPI", BMEASIFL == "Y") %>%
   mutate(

graphs/RFG2/setup.qmd

@@ -13,15 +13,15 @@ In detail the differences to [RFG1](../../graphs/RFG1/rfg01.qmd) are the followi
 -   The `extract_rsp_subgroups()` and `tabulate_rsp_subgroups()` functions only allow specification of a single treatment arm variable, while the `extract_rsp_biomarkers()` and `tabulate_rsp_biomarkers()` allow to look at multiple continuous biomarker variables at once.
 -   In addition to the treatment arms, the use of `extract_rsp_subgroups()` and `tabulate_rsp_subgroups()` functions can be extended to other binary variables. For example, we could define the binarized `ARM` variable as `AGE>=65` vs. `AGE<65` and then look at the odds ratios across subgroups. For the `extract_rsp_biomarkers()` and `tabulate_rsp_biomarkers()` functions, we could use the original continuous biomarker variable `AGE`, and then look at the estimated effect across subgroups.
 
-Similarly like in [RFG1](../../graphs/RFG1/rfg01.qmd), we will use the `synthetic_cdisc_data$adrs` data set from the `scda` package.
+Similarly like in [RFG1](../../graphs/RFG1/rfg01.qmd), we will use the `cadrs` data set from the `random.cdisc.data` package.
 Here we just filter for the Best Confirmed Overall Response by Investigator and patients with measurable disease at baseline, and define a new variable `COMPRESP` to include complete responses only.
 
 ```{r, message = FALSE}
 library(tern)
 library(dplyr)
 library(hermes)
 
-adrs <- scda::synthetic_cdisc_data("latest")$adrs
+adrs <- random.cdisc.data::cadrs
 
 adrs_f <- adrs %>%
   df_explicit_na() %>%

graphs/RG1/setup.qmd

@@ -4,7 +4,7 @@ exclude-listing: true
 
 ## Setup
 
-We will use the `synthetic_cdisc_data$adrs` data set from the `scda` package to create the response plots.
+We will use the `cadrs` data set from the `random.cdisc.data` package to create the response plots.
 We transform the response variable into an `ordered` factor to ensure that the response labels are ordered correctly and a sequential color scheme is used in the graph.
 We select Best Confirmed Overall Response by Investigator to evaluate response.
 Finally, we select patients with measurable disease at baseline `(BMEASIFL == "Y")` as response evaluable patients.
@@ -16,7 +16,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adrs <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adrs %>%
+adrs <- random.cdisc.data::cadrs %>%
   df_explicit_na() %>%
   mutate(AVALC = ordered(AVALC, levels = c("<Missing>", "NE", "PD", "SD", "PR", "CR"))) %>%
   filter(PARAMCD == "BESRSPI", BMEASIFL == "Y")

graphs/RG2/setup.qmd

@@ -13,7 +13,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adrs <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adrs %>%
+adrs <- random.cdisc.data::cadrs %>%
   df_explicit_na() %>%
   mutate(AVALC = ordered(AVALC, levels = c("<Missing>", "NE", "PD", "SD", "PR", "CR"))) %>%
   filter(PARAMCD == "BESRSPI", BMEASIFL == "Y")

graphs/RG3/setup.qmd

@@ -13,7 +13,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adrs <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adrs %>%
+adrs <- random.cdisc.data::cadrs %>%
   df_explicit_na() %>%
   mutate(
     AVALC = ordered(AVALC, levels = c("<Missing>", "NE", "PD", "SD", "PR", "CR")),

graphs/SFG1/setup.qmd

@@ -4,7 +4,7 @@ exclude-listing: true
 
 ## Setup
 
-We will use the `synthetic_cdisc_data$adtte` data set from the `scda` package to create the survival forest graph.
+We will use the `cadtte` data set from the `random.cdisc.data` package to create the survival forest graph.
 We start by filtering the `adtte` data set for the overall survival observations, converting time of overall survival to months, creating a new variable for event information, binarizing the `ARM` variable and creating a binned age variable by using the function `cut_quantile_bins()`.
 Note that we do not include the boundaries `0` and `1` in the corresponding cutoffs vector `AGE_probs`, but only the true cutoff probabilities to use (here `0.5`, i.e. the median).
 We restrict the analysis of the biomarker variables `BMRKR1` and `BMRKR2` to the BEP by setting them as missing for the non-BEP.
@@ -17,7 +17,7 @@ library(dplyr)
 
 AGE_probs <- 0.5
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS"

graphs/SFG2/setup.qmd

@@ -13,7 +13,7 @@ library(dplyr)
 
 BMRKR1_probs <- c(0.25, 0.5, 0.75)
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS"

graphs/SFG3/setup.qmd

@@ -10,7 +10,7 @@ We prepare the data similarly as in [SFG1](../../graphs/SFG1/sfg01.qmd), focusin
 library(tern)
 library(dplyr)
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS",

graphs/SFG3/sfg03a.qmd

@@ -12,11 +12,11 @@ categories: [SFG]
 ## Plot
 
 We prepare the data similarly as in [SFG3](sfg03.qmd).
-Additionally we are filtering `synthetic_cdisc_data$adtte` to keep only two categories for the `SEX` variable (otherwise we would not be able to do the forest plot), and we are keeping all ITT patients.
+Additionally we are filtering `random.cdisc.data::cadtte` to keep only two categories for the `SEX` variable (otherwise we would not be able to do the forest plot), and we are keeping all ITT patients.
 We then tabulate statistics to be able to use them as an input for the forest plot.
 
 ```{r}
-adtte_mf <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte_mf <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS",

graphs/SFG5/setup.qmd

@@ -12,7 +12,7 @@ library(dplyr)
 
 AGE_probs <- 0.5
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS"

graphs/SFG6/setup.qmd

@@ -13,14 +13,14 @@ In detail the differences to the other survival forest graphs ([SFG1](../../grap
 -   The `extract_survival_subgroups()` and `tabulate_survival_subgroups()` functions only allow specification of a single treatment arm variable, while the `extract_survival_biomarkers()` and `tabulate_survival_biomarkers()` allow to look at multiple continuous biomarker variables at once.
 -   In addition to the treatment arms, the use of `extract_survival_subgroups()` and `tabulate_survival_subgroups()` functions can be extended to other binary variables, as done in [SFG3](../../graphs/SFG3/sfg03.qmd) and [SFG4](../../graphs/sfg04.qmd). For example, we could define the binarized `ARM` variable as `AGE>=65` vs. `AGE<65` and then look at the odds ratios across subgroups. For the `extract_survival_biomarkers()` and `tabulate_survival_biomarkers()` functions, we could use the original continuous biomarker variable `AGE`, and then look at the estimated effect across subgroups.
 
-Similarly like in [SFG3](../../graphs/SFG3/sfg03.qmd), we will use the `synthetic_cdisc_data$adtte` data set from the `scda` package.
+Similarly like in [SFG3](../../graphs/SFG3/sfg03.qmd), we will use the `cadtte` data set from the `random.cdisc.data` package.
 Here we just filter for the overall survival outcome in a single arm in the biomarker evaluable population.
 
 ```{r, message = FALSE}
 library(tern)
 library(dplyr)
 
-adtte_f <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte_f <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS",

graphs/ag01.qmd

@@ -31,7 +31,7 @@ library(tidyr)
 library(dplyr)
 library(ggplot2.utils)
 
-adsl <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adsl %>%
+adsl <- random.cdisc.data::cadsl %>%
   df_explicit_na() %>%
   drop_na(BMRKR1)
 

graphs/dg02.qmd

@@ -11,7 +11,7 @@ categories: [DG]
 
 The graph below plots the distribution of a biomarker variable (on a continuous scale) as a boxplot by one or more categorical clinical variables with overlaid points.
 
-We will use the `synthetic_cdisc_data$adsl` data set from the `scda` package to illustrate the graph and will select the biomarker evaluable population with `BEP01FL`.
+We will use the `cadsl` data set from the `random.cdisc.data` package to illustrate the graph and will select the biomarker evaluable population with `BEP01FL`.
 The column `BMRKR1` contains the biomarker values on a continuous scale.
 We will use `STRATA2` and `ARM` as categorical clinical variables.
 
@@ -20,7 +20,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adsl <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adsl %>%
+adsl <- random.cdisc.data::cadsl %>%
   df_explicit_na() %>%
   filter(BEP01FL == "Y")
 ```

graphs/dg04.qmd

@@ -11,15 +11,15 @@ categories: [DG]
 
 The graph below plots two continuous (biomarker) variables against each other.
 
-We will use the `synthetic_cdisc_data$adsl` data set from the `scda` package to illustrate the graph and will select the biomarker evaluable population with `BEP01FL`.
+We will use the `cadsl` data set from the `random.cdisc.data` package to illustrate the graph and will select the biomarker evaluable population with `BEP01FL`.
 The columns `AGE` and `BMRKR1` contain the biomarker values of interest on a continuous scale.
 
 ```{r, message = FALSE}
 library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adsl <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adsl %>%
+adsl <- random.cdisc.data::cadsl %>%
   df_explicit_na() %>%
   filter(BEP01FL == "Y")
 ```

graphs/kg03.qmd

@@ -16,7 +16,7 @@ The difference is that here we use the categorical biomarker variable `BMRKR2` a
 library(tern)
 library(dplyr)
 
-adtte_arm_bep <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte_arm_bep <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(PARAMCD == "OS", ARM == "A: Drug X", BEP01FL == "Y") %>%
   mutate(is_event = CNSR == 0)

graphs/rfg03.qmd

@@ -19,7 +19,7 @@ library(tern)
 library(ggplot2.utils)
 library(dplyr)
 
-adrs <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adrs %>%
+adrs <- random.cdisc.data::cadrs %>%
   df_explicit_na() %>%
   filter(PARAMCD == "BESRSPI", BMEASIFL == "Y") %>%
   mutate(

graphs/rnag09.qmd

@@ -22,7 +22,7 @@ We define an event indicator variable, transform the time to months and filter d
 library(tern)
 library(dplyr)
 
-adtte_f <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte_f <- random.cdisc.data::cadtte %>%
   dplyr::mutate(
     is_event = .data$CNSR == 0,
     AVAL = day2month(.data$AVAL),
@@ -88,7 +88,7 @@ See [KG1](../graphs/KG1/kg01.qmd) to [KG5](../graphs/KG5/kg05.qmd) for additiona
 
 ## Teal Module for Kaplan-Meier Graph
 
-We start by importing a `MultiAssayExperiment` and sample `ADTTE` data; here we use the example `multi_assay_experiment` available in `hermes` and example `ADTTE` data from `scda`.
+We start by importing a `MultiAssayExperiment` and sample `ADTTE` data; here we use the example `multi_assay_experiment` available in `hermes` and example `ADTTE` data from `random.cdisc.data`.
 We can then use the provided teal module `tm_g_km` to include the corresponding interactive Kaplan-Meier analysis in our teal app.
 Note that by default the `counts` assay is excluded via the `exclude_assays` argument, but we can include it by just saying that we don't want to exclude any assays.
 In case that we have different non-standard column names in our ADTTE data set we could also specify them via the `adtte_vars` argument, see the documentation `?tm_g_km` for more details.
@@ -98,11 +98,11 @@ library(teal.modules.hermes)
 
 data <- teal_data()
 data <- within(data, {
-  library(scda)
+  library(random.cdisc.data)
   library(dplyr)
   library(hermes)
   MAE <- multi_assay_experiment
-  ADTTE <- synthetic_cdisc_dataset("latest", "adtte") %>%
+  ADTTE <- random.cdisc.data::cadtte %>%
     mutate(is_event = .data$CNSR == 0)
 })
 datanames(data) <- c("MAE", "ADTTE")

graphs/rnag10.qmd

@@ -21,7 +21,7 @@ We define an event indicator variable, transform the time to months and filter d
 ```{r, message = FALSE}
 library(tern)
 
-adtte_f <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte_f <- random.cdisc.data::cadtte %>%
   dplyr::mutate(
     is_event = .data$CNSR == 0,
     AVAL = day2month(.data$AVAL),
@@ -99,7 +99,7 @@ See [SFG01](../graphs/SFG1/sfg01.qmd) to [SFG05](../graphs/SFG5/sfg05.qmd) for a
 
 ## Teal Module for Survival Forest Graph
 
-We start by importing a `MultiAssayExperiment` and sample `ADTTE` data; here we use the example `multi_assay_experiment` available in `hermes` and example `ADTTE` data from `scda`.
+We start by importing a `MultiAssayExperiment` and sample `ADTTE` data; here we use the example `multi_assay_experiment` available in `hermes` and example `ADTTE` data from `random.cdisc.data`.
 We can then use the provided teal module `tm_g_forest_tte` to include the corresponding interactive survival forest analysis in our teal app.
 In case that we have different non-standard column names in our ADTTE data set we could also specify them via the `adtte_vars` argument, see the documentation `?tm_g_forest_tte` for more details.
 
@@ -108,11 +108,11 @@ library(teal.modules.hermes)
 
 data <- teal_data()
 data <- within(data, {
-  library(scda)
+  library(random.cdisc.data)
   library(dplyr)
   library(hermes)
   MAE <- multi_assay_experiment
-  ADTTE <- synthetic_cdisc_dataset("latest", "adtte") %>%
+  ADTTE <- random.cdisc.data::cadtte %>%
     mutate(is_event = .data$CNSR == 0)
 })
 datanames(data) <- c("MAE", "ADTTE")

graphs/sfg04.qmd

@@ -15,7 +15,7 @@ We prepare the data similarly as in [SFG1](../graphs/SFG1/sfg01.qmd).
 library(tern)
 library(dplyr)
 
-adtte <- scda::synthetic_cdisc_data("rcd_2022_06_27")$adtte %>%
+adtte <- random.cdisc.data::cadtte %>%
   df_explicit_na() %>%
   filter(
     PARAMCD == "OS",