Rd modifications

Former-commit-id: 386607160b2e87fada9c19a6f656f3a655716041
Former-commit-id: da0304e2fa91100465d8296c021f440e3ae1bb1f
Former-commit-id: 5f47cb2441719006771c95b31b7a079c2094971a

dev/flat_full.Rmd

@@ -599,17 +599,17 @@ test_that("isValidMock_obj checks if the provided object is a valid mock object"
 #'
 #' @param list_var Either c() or output of init_variable
 #' @param name Variable name
-#' @param mu Either a numeric value or a numeric vector (of length = level)
-#' @param sd Either numeric value or NA
+#' @param sd Either numeric value or NA. Use to specify range of effect sizes.
 #' @param level Numeric value to specify the number of levels to simulate
+#' @param mu Either a numeric value or a numeric vector (of length = level). Default : 0. Not recommended to modify.
 #'
 #' @return
 #' A list with initialized variables
 #' @export
 #'
 #' @examples
-#' init_variable(name = "my_varA", mu = 2, sd = 9, level = 200)
-init_variable <- function(list_var = c(), name = "myVariable", mu = c(2,3), sd = NA, level = NA) {
+#' init_variable(name = "my_varA", sd = 9, level = 200)
+init_variable <- function(list_var = c(), name = "myVariable", sd = NA, level = NA, mu = 0) {
   
   name <- clean_variable_name(name)
   
@@ -795,18 +795,19 @@ build_sub_obj_return_to_user <- function(level, metaData, effectsGivenByUser, co
 #'
 #' @param list_var A list of variables (already initialized)
 #' @param between_var A vector of variable names to include in the interaction
-#' @param mu Either a numeric value or a numeric vector (of length = level)
-#' @param sd Either numeric value or NA
+#' @param sd Either numeric value or NA. Use to specify range of effect sizes.
+#' @param mu Either a numeric value or a numeric vector (of length = level). Default : 0. Not recommended to modify.
+
 #'
 #' @return
 #' A list with initialized interaction
 #' @export
 #'
 #' @examples
-#' init_variable(name = "myvarA", mu = 2, sd = 3, level = 200) %>%
-#' init_variable(name = "myvarB", mu = 1, sd = 0.2, level = 2 ) %>%
-#' add_interaction(between_var = c("myvarA", "myvarB"), mu = 3, sd = 2)
-add_interaction <- function(list_var, between_var, mu, sd = NA) {
+#' init_variable(name = "myvarA", sd = 3, level = 200) %>%
+#' init_variable(name = "myvarB", sd = 0.2, level = 2 ) %>%
+#' add_interaction(between_var = c("myvarA", "myvarB"), sd = 2)
+add_interaction <- function(list_var, between_var, sd = NA, mu = 0) {
   name_interaction <- paste(between_var, collapse = ":")
   check_input2interaction(name_interaction, list_var, between_var, mu, sd)
   
@@ -7819,12 +7820,9 @@ evaluation_report <- function(mock_obj, list_gene = NULL,
         list_tmb <- list_tmb[list_gene]
   }
   
-  
   ## -- eval data
   eval_data <- get_eval_data(list_tmb, dds, mock_obj, coeff_threshold, alt_hypothesis)
-  
   ## -- identity plot
-  #identity_data <- rbind_model_params_and_dispersion(eval_data)
   params_identity_eval <- eval_identityTerm( eval_data$modelparams, palette_color, palette_shape )
   dispersion_identity_eval <- eval_identityTerm(eval_data$modeldispersion, palette_color, palette_shape)
   
@@ -7834,6 +7832,12 @@ evaluation_report <- function(mock_obj, list_gene = NULL,
     eval_data2metrics <- eval_data$modelparams
   }
 
+  ## aggregate RMSE + R2
+  rmse_modelparams <- compute_rmse(eval_data2metrics, grouping_by = "from")
+  rsquare_modelparams <- compute_rsquare(eval_data2metrics, grouping_by = "from")
+  rsquare_modelparams$from <- NULL
+  aggregate_metrics <- cbind(rmse_modelparams, rsquare_modelparams)
+  
   
   ## -- counts plot
   counts_violinplot <- counts_plot(mock_obj)
@@ -7848,9 +7852,11 @@ evaluation_report <- function(mock_obj, list_gene = NULL,
     return(
           list(
                 data = eval_data, 
-                identity = list( params = params_identity_eval$p,
+                identity = list(params = params_identity_eval$p,
                                 dispersion = dispersion_identity_eval$p ), 
-                counts = counts_violinplot
+                counts = counts_violinplot,
+                performances = list(byparams = rbind(dispersion_identity_eval$R2, params_identity_eval$R2),
+                                    aggregate = aggregate_metrics )
                 
           ))
       
@@ -7870,7 +7876,8 @@ evaluation_report <- function(mock_obj, list_gene = NULL,
   ## -- acc, recall, sensib, speci, ...
   metrics_obj <- get_ml_metrics_obj(eval_data2metrics, alpha_risk )
   ## -- merge all metrics in one obj
-  model_perf_obj <- get_performances_metrics_obj( r2_params = params_identity_eval$R2,                        
+  model_perf_obj <- get_performances_metrics_obj( r2_params = params_identity_eval$R2,   
+                                                  r2_agg = aggregate_metrics, 
                                                   dispersion_identity_eval$R2, 
                                                   pr_curve_obj,
                                                   roc_curve_obj,
@@ -7910,15 +7917,16 @@ evaluation_report <- function(mock_obj, list_gene = NULL,
 #' The function generates separate data frames for metric values by parameter value and for the
 #' aggregated metric values.
 #' 
-#' @param r2_params R-squared values from model parameters evaluation object.
-#' @param r2_dispersion R-squared values from dispersion evaluation object.
+#' @param r2_params R-squared and RMSE values from model parameters evaluation object.
+#' @param r2_agg R-squared and RMSE values aggregated.
+#' @param r2_dispersion R-squared and RMSE values from dispersion evaluation object.
 #' @param pr_obj PR object generated from evaluation report.
 #' @param roc_obj ROC object generated from evaluation report.
 #' @param ml_metrics_obj Machine learning performance metrics object.
 #' 
 #' @return A list containing separate data frames for by-parameter and aggregated metric values.
 #' @export 
-get_performances_metrics_obj <- function(r2_params , r2_dispersion,
+get_performances_metrics_obj <- function(r2_params , r2_agg ,r2_dispersion,
                                          pr_obj, roc_obj, ml_metrics_obj ){
   ## -- by params  
   auc_mtrics_params <- join_dtf(pr_obj$byparams$pr_auc , roc_obj$byparams$roc_auc,
@@ -7933,6 +7941,8 @@ get_performances_metrics_obj <- function(r2_params , r2_dispersion,
                       k1 = c("from"), k2 = c("from"))
   metrics_agg <- join_dtf(auc_mtrics_agg , ml_metrics_obj$aggregate,
                              k1 = c("from"), k2 = c("from"))
+  metrics_agg <- join_dtf(metrics_agg, r2_agg,
+                          k1 = c("from"), k2 = c("from"))
   return(list(byparams = metrics_params, aggregate = metrics_agg ))  
 }
 
@@ -8064,8 +8074,6 @@ test_that("isValidEvalInput with valid input", {
 
 
 test_that("evaluation_report returns correct output", {
-  
-    
   N_GENES <- 100
   MAX_REPLICATES <- 5
   MIN_REPLICATES <- 5
@@ -8127,8 +8135,12 @@ test_that("get_performances_metrics_obj returns correct output", {
                                                 recall = c(0.09, 0.88)),
                          aggregate = data.frame(from = c("Glm", "Hglm"),
                                                 accuracy = c(0.7, 0.6), recall = c(0.1, 0.8)))
+  r2_agg <- data.frame(from = c("Glm", "Hglm"),
+                          RMSE = c(0.22, 0.55),
+                          R2 = c(0.9, 0.7))
+  
   # Call the function
-  result <- get_performances_metrics_obj(r2_params, r2_dispersion,
+  result <- get_performances_metrics_obj(r2_params, r2_agg, r2_dispersion,
                                           pr_obj, roc_obj, ml_metrics_obj)
   
   # Test the output
@@ -8138,7 +8150,7 @@ test_that("get_performances_metrics_obj returns correct output", {
   expect_equal(nrow(result$byparams), 6)
   expect_equal(ncol(result$byparams), 7)
   expect_equal(nrow(result$aggregate), 2)
-  expect_equal(ncol(result$aggregate), 5)
+  expect_equal(ncol(result$aggregate), 7)
 })
 
 
@@ -8193,7 +8205,6 @@ test_that("get_ml_metrics_obj returns correct output", {
 
 
 
-
 ```
 
 
@@ -8565,8 +8576,8 @@ test_that("get_eval_data returns correct output", {
 #' MAX_REPLICATES = 5
 #' MIN_REPLICATES = 5
 #' ## --init variable
-#' input_var_list <- init_variable( name = "genotype", mu = 12, sd = 0.1, level = 3) %>%
-#'                    init_variable(name = "environment", mu = c(0,1), NA , level = 2) 
+#' input_var_list <- init_variable( name = "genotype", sd = 0.1, level = 3) %>%
+#'                    init_variable(name = "environment", NA , level = 2) 
 #'
 #' mock_data <- mock_rnaseq(input_var_list, N_GENES, MIN_REPLICATES, MAX_REPLICATES)
 #' dds <- DESeq2::DESeqDataSetFromMatrix(mock_data$counts , 
@@ -9136,7 +9147,7 @@ test_that("RMSE function calculates correct RMSE values", {
 #' N_GENES = 100
 #' MAX_REPLICATES = 5
 #' MIN_REPLICATES = 5
-#' input_var_list <- init_variable(name = "varA", mu = 10, sd = 0.1, level = 3)
+#' input_var_list <- init_variable(name = "varA", sd = 0.1, level = 3)
 #' mock_data <- mock_rnaseq(input_var_list, N_GENES,
 #'                         min_replicates = MIN_REPLICATES, 
 #'                         max_replicates = MAX_REPLICATES)
@@ -9711,7 +9722,8 @@ isValidEval_report <- function(obj){
     stop("All elements in 'obj' are NULL")
   }
   ## level 1 
-  expected_names <- c("data", "identity", "precision_recall", "roc", "counts", "performances")
+  expected_names <- c("data", "identity", "counts", "performances")
+  ## N.B : "precision_recall", "roc" names are not mandatory
   if (!all(expected_names %in% names(obj))) {
     stop(message_err)
   }
@@ -9782,15 +9794,21 @@ export_evaluation_report <- function(eval_report_obj, outfolder, plot_format = "
 #' @importFrom ggplot2 ggsave
 #' @export
 export_eval_plots <- function(eval_report_obj, extension, ...){
+  
   dir.create("plots", showWarnings = FALSE)
   setwd("./plots")
+  
   ggplot2::ggsave(filename = paste("identity_modelparams" , extension, sep = "."), eval_report_obj$identity$params, ...)
   ggplot2::ggsave(filename = paste("identity_modeldispersion" , extension, sep = "."), eval_report_obj$identity$dispersion )
-  ggplot2::ggsave(filename = paste("precision_recall_byparams" , extension, sep = "."), eval_report_obj$precision_recall$params )
-  ggplot2::ggsave(filename = paste("precision_recall_aggregated" , extension, sep = "."), eval_report_obj$precision_recall$aggregate)
-  ggplot2::ggsave(filename = paste("roc_byparams" , extension, sep = "."), eval_report_obj$roc$params )
-  ggplot2::ggsave(filename = paste("roc_aggregated" , extension, sep = "."), eval_report_obj$roc$aggregate )
   ggplot2::ggsave(filename = paste("genes_expression" , extension, sep = "."), eval_report_obj$counts)
+  
+  if (all(c("roc", "precision_recall") %in% names(eval_report_obj))){
+      ggplot2::ggsave(filename = paste("precision_recall_byparams" , extension, sep = "."), eval_report_obj$precision_recall$params )
+      ggplot2::ggsave(filename = paste("precision_recall_aggregated" , extension, sep = "."), eval_report_obj$precision_recall$aggregate)
+      ggplot2::ggsave(filename = paste("roc_byparams" , extension, sep = "."), eval_report_obj$roc$params )
+      ggplot2::ggsave(filename = paste("roc_aggregated" , extension, sep = "."), eval_report_obj$roc$aggregate )
+  }
+
 }
 
 

man/add_interaction.Rd

@@ -4,16 +4,16 @@
 \alias{add_interaction}
 \title{Add interaction}
 \usage{
-add_interaction(list_var, between_var, mu, sd = NA)
+add_interaction(list_var, between_var, sd = NA, mu = 0)
 }
 \arguments{
 \item{list_var}{A list of variables (already initialized)}
 
 \item{between_var}{A vector of variable names to include in the interaction}
 
-\item{mu}{Either a numeric value or a numeric vector (of length = level)}
+\item{sd}{Either numeric value or NA. Use to specify range of effect sizes.}
 
-\item{sd}{Either numeric value or NA}
+\item{mu}{Either a numeric value or a numeric vector (of length = level). Default : 0. Not recommended to modify.}
 }
 \value{
 A list with initialized interaction
@@ -22,7 +22,7 @@ A list with initialized interaction
 Add interaction
 }
 \examples{
-init_variable(name = "myvarA", mu = 2, sd = 3, level = 200) \%>\%
-init_variable(name = "myvarB", mu = 1, sd = 0.2, level = 2 ) \%>\%
-add_interaction(between_var = c("myvarA", "myvarB"), mu = 3, sd = 2)
+init_variable(name = "myvarA", sd = 3, level = 200) \%>\%
+init_variable(name = "myvarB", sd = 0.2, level = 2 ) \%>\%
+add_interaction(between_var = c("myvarA", "myvarB"), sd = 2)
 }

man/get_performances_metrics_obj.Rd

@@ -6,6 +6,7 @@
 \usage{
 get_performances_metrics_obj(
   r2_params,
+  r2_agg,
   r2_dispersion,
   pr_obj,
   roc_obj,
@@ -13,9 +14,11 @@ get_performances_metrics_obj(
 )
 }
 \arguments{
-\item{r2_params}{R-squared values from model parameters evaluation object.}
+\item{r2_params}{R-squared and RMSE values from model parameters evaluation object.}
 
-\item{r2_dispersion}{R-squared values from dispersion evaluation object.}
+\item{r2_agg}{R-squared and RMSE values aggregated.}
+
+\item{r2_dispersion}{R-squared and RMSE values from dispersion evaluation object.}
 
 \item{pr_obj}{PR object generated from evaluation report.}
 

man/init_variable.Rd

@@ -4,24 +4,18 @@
 \alias{init_variable}
 \title{Initialize variable}
 \usage{
-init_variable(
-  list_var = c(),
-  name = "myVariable",
-  mu = c(2, 3),
-  sd = NA,
-  level = NA
-)
+init_variable(list_var = c(), name = "myVariable", sd = NA, level = NA, mu = 0)
 }
 \arguments{
 \item{list_var}{Either c() or output of init_variable}
 
 \item{name}{Variable name}
 
-\item{mu}{Either a numeric value or a numeric vector (of length = level)}
-
-\item{sd}{Either numeric value or NA}
+\item{sd}{Either numeric value or NA. Use to specify range of effect sizes.}
 
 \item{level}{Numeric value to specify the number of levels to simulate}
+
+\item{mu}{Either a numeric value or a numeric vector (of length = level). Default : 0. Not recommended to modify.}
 }
 \value{
 A list with initialized variables
@@ -30,5 +24,5 @@ A list with initialized variables
 Initialize variable
 }
 \examples{
-init_variable(name = "my_varA", mu = 2, sd = 9, level = 200)
+init_variable(name = "my_varA", sd = 9, level = 200)
 }

man/subsetGenes.Rd

@@ -24,7 +24,7 @@ This function filters and adjusts genomic data within the Roxygeb project, based
 N_GENES = 100
 MAX_REPLICATES = 5
 MIN_REPLICATES = 5
-input_var_list <- init_variable(name = "varA", mu = 10, sd = 0.1, level = 3)
+input_var_list <- init_variable(name = "varA", sd = 0.1, level = 3)
 mock_data <- mock_rnaseq(input_var_list, N_GENES,
                         min_replicates = MIN_REPLICATES, 
                         max_replicates = MAX_REPLICATES)

man/wrap_dds.Rd

@@ -31,8 +31,8 @@ N_GENES = 100
 MAX_REPLICATES = 5
 MIN_REPLICATES = 5
 ## --init variable
-input_var_list <- init_variable( name = "genotype", mu = 12, sd = 0.1, level = 3) \%>\%
-                   init_variable(name = "environment", mu = c(0,1), NA , level = 2) 
+input_var_list <- init_variable( name = "genotype", sd = 0.1, level = 3) \%>\%
+                   init_variable(name = "environment", NA , level = 2) 
 
 mock_data <- mock_rnaseq(input_var_list, N_GENES, MIN_REPLICATES, MAX_REPLICATES)
 dds <- DESeq2::DESeqDataSetFromMatrix(mock_data$counts ,