change functions to force all data arguments to be treated as data frames

ritamtbsilva · ritamtbsilva · commit e390568bddc5 · 2025-07-14T14:09:27.000+01:00
diff --git a/NEWS.md b/NEWS.md
@@ -1,5 +1,7 @@
 # markeR 0.9.X (?? ??, ????)
+
 * Added `VisualiseIndividualGenes()` wrapper to unify individual gene visualisation functions (`ExpressionHeatmap`, `ROCandAUCplot`, etc.) under a single, user-friendly interface.
+* Ensured all data arguments are data frames for consistency across functions.
 
 # markeR 0.9.4 (09 Jul, 2025)
 
diff --git a/R/CalculateScores.R b/R/CalculateScores.R
@@ -92,7 +92,7 @@
 #'
 #' @export
 CalculateScores <- function(data, metadata, gene_sets, method = c("ssGSEA", "logmedian","ranking", "all")) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   method <- match.arg(method)  # Validate method input
 
   if (!is.data.frame(data)) stop("Error: data must be a data-frame")
diff --git a/R/CalculateScores_Ranking.R b/R/CalculateScores_Ranking.R
@@ -45,7 +45,7 @@
 #'}
 #' @keywords internal
 CalculateScores_Ranking <- function(data, metadata = NULL, gene_sets) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   ResultsList <- list()
 
   if (!is.data.frame(data)) stop("Error: data must be a data frame")
@@ -147,7 +147,7 @@ CalculateScores_Ranking <- function(data, metadata = NULL, gene_sets) {
 #'}
 #' @keywords internal
 getRanking <- function(data, sample, geneset) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   expressiongene <- data[, sample]  # Isolate one sample and get the expression of all genes
   names(expressiongene) <- row.names(data)  # Name vector
   expressiongene <- expressiongene[order(expressiongene, decreasing = FALSE)]  # Order from least to most expressed
diff --git a/R/CalculateScores_logmedian.R b/R/CalculateScores_logmedian.R
@@ -35,6 +35,7 @@
 #' }
 #' @keywords internal
 CalculateScores_logmedian <- function(data, metadata = NULL, gene_sets) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   ResultsList <- list()
 
   for (sig in names(gene_sets)) {
@@ -83,6 +84,7 @@ CalculateScores_logmedian <- function(data, metadata = NULL, gene_sets) {
 #' @return A named vector with log-median-centered scores per sample.
 #' @keywords internal
 calculateScore_logmedian_bidirectional <- function(data, signature) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   signature[, 2] <- as.numeric(signature[, 2])
 
   if (!all(unique(signature[, 2]) %in% c(-1, 1))) {
@@ -122,6 +124,7 @@ calculateScore_logmedian_bidirectional <- function(data, signature) {
 #' @return A named vector with log-median-centered scores per sample.
 #' @keywords internal
 calculateScore_logmedian_unidirectional <- function(data, signature) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   if (is.data.frame(signature)) signature <- as.vector(signature[, 1])
 
   data_subset <- na.omit(subset(log2(data + 1), row.names(data) %in% signature))
diff --git a/R/CalculateScores_ssGSEA.R b/R/CalculateScores_ssGSEA.R
@@ -21,6 +21,7 @@
 #' @keywords internal
 
 CalculateScores_ssGSEA <- function(data, metadata = NULL, gene_sets) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   ResultsList <- list()
 
   for (sig in names(gene_sets)) {
@@ -89,6 +90,7 @@ CalculateScores_ssGSEA <- function(data, metadata = NULL, gene_sets) {
 #'}
 #' @keywords internal
 CalculateScores_ssGSEA_unidirectional <- function(data, signature) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   ResultsList <- list()
 
   siglist <- list(signature)
@@ -150,6 +152,7 @@ CalculateScores_ssGSEA_unidirectional <- function(data, signature) {
 #'}
 #' @keywords internal
 CalculateScores_ssGSEA_bidirectional <- function(data, signature) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   ResultsList <- list()
 
   mtx <- log2(data)
diff --git a/R/CohenD_IndividualGenes.R b/R/CohenD_IndividualGenes.R
@@ -97,7 +97,7 @@ CohenD_IndividualGenes <- function(data, metadata,
                           title = NULL,
                           titlesize = 16,
                           params = list()) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # If genes are not specified, use all available genes
   if (is.null(genes)) {
     genes <- rownames(data)
diff --git a/R/CorrelationHeatmap.R b/R/CorrelationHeatmap.R
@@ -116,7 +116,7 @@ CorrelationHeatmap <- function(data, metadata = NULL, genes, separate.by = NULL,
                                show_row_names=TRUE,
                                show_column_names=TRUE) {
 
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # Choose legend position: "side" (vertical) or "top" (horizontal)
   legend_position <- match.arg(legend_position)
   method <- match.arg(method)
diff --git a/R/ExpressionHeatmap.R b/R/ExpressionHeatmap.R
@@ -88,7 +88,7 @@ ExpressionHeatmap <- function(data, metadata = NULL, genes, annotate.by = NULL,
                               legend_position = c("top", "right", "bottom"),
                               show_row_names=TRUE,
                               show_column_names=FALSE) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # Ensure the scale_position and legend_position arguments are matched correctly
   scale_position <- match.arg(scale_position)
   legend_position <- match.arg(legend_position)
diff --git a/R/FPR_Simulation.R b/R/FPR_Simulation.R
@@ -91,7 +91,7 @@
 FPR_Simulation <- function(data, metadata, original_signatures, Variable, gene_list = NULL, number_of_sims=10, title=NULL,
                            widthTitle = 30, titlesize = 12,  pointSize = 2,
                            labsize = 10,mode = c( "none","simple","medium","extensive"), ColorValues=NULL, ncol=NULL, nrow=NULL) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   if (is.null(gene_list)) gene_list <- row.names(data)
 
   methods <- c("ssGSEA", "logmedian", "ranking")
diff --git a/R/GSEA_VariableAssociation.R b/R/GSEA_VariableAssociation.R
@@ -82,6 +82,7 @@
 #'
 #' @keywords internal
 GSEA_VariableAssociation <- function(data, metadata, cols, stat=NULL, mode=c("simple","medium","extensive"), gene_set,nonsignif_color = "grey", signif_color = "red", saturation_value=NULL,sig_threshold = 0.05, widthlabels=18, labsize=10, titlesize=14, pointSize=5, ignore_NAs = FALSE, printplt =TRUE) {
+  data <- as.data.frame(data) # Ensure data is a data frame
   mode <- match.arg(mode)
   metadata <- metadata[, cols %in% colnames(metadata), drop = FALSE]
 
diff --git a/R/Heatmap_Cohen.R b/R/Heatmap_Cohen.R
@@ -187,7 +187,7 @@ Heatmap_Cohen <- function(cohenlist, nrow = NULL, ncol = NULL, limits = NULL, wi
 #'
 #' @keywords internal
 CohenD_allConditions <- function(data, metadata, gene_sets, variable, mode = c("simple","medium","extensive")) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # Step 1: Check if variable exists in metadata
   if (!variable %in% colnames(metadata)) {
     stop(paste("Error: Variable", variable, "not found in metadata."))
@@ -444,7 +444,7 @@ flatten_results <- function(nested_list) {
 #'
 #' @keywords internal
 CohenF_allConditions <- function(data, metadata, gene_sets, variable ) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # Step 1: Check if variable exists in metadata
   if (!variable %in% colnames(metadata)) {
     stop(paste("Error: Variable", variable, "not found in metadata."))
diff --git a/R/PlotScores.R b/R/PlotScores.R
@@ -202,7 +202,7 @@ PlotScores <- function(data, metadata, gene_sets,
                        widthTitle = 20, titlesize = 12, limits = NULL, legend_nrow = NULL, pointSize = 4,
                        xlab = NULL, labsize = 10, compute_cohen=TRUE, cond_cohend = NULL, pvalcalc = FALSE, mode = c("simple","medium","extensive"),
                        widthlegend=22, sig_threshold=0.05, cohen_threshold=0.5, colorPalette="Set3", cor=c("pearson","spearman","kendall")) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   method <- match.arg(method)
   mode <- match.arg(mode)
   cor <- match.arg(cor)
@@ -378,7 +378,7 @@ PlotScores_Categorical <- function(data, metadata, gene_sets,
                                    widthTitle = 10, titlesize = 12, limits = NULL, legend_nrow = NULL, pointSize = 2,
                                    xlab = NULL, labsize = 10, compute_cohen=TRUE, cond_cohend = NULL, pvalcalc = FALSE, mode = c("simple","medium","extensive"),
                                    widthlegend=22, cohen_threshold=0.6, colorPalette="Set3") {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   method <- match.arg(method)
 
   ResultsList <- CalculateScores(data = data,
@@ -745,7 +745,7 @@ PlotScores_Numeric <- function(data,
                                pvalcalc = FALSE,
                                colorPalette = "Set3",
                                cor = c("pearson","spearman","kendall")) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   method <- match.arg(method)
 
 
diff --git a/R/ROCAUC_Scores.R b/R/ROCAUC_Scores.R
@@ -26,7 +26,7 @@ ROCAUC_Scores_Calculate <- function(data, metadata, gene_sets, method = c("logme
 
   method <- match.arg(method)
   mode <- match.arg(mode)
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   if (method != "all") {
     listScores <- list(CalculateScores(data = data, metadata = metadata, gene_sets = gene_sets, method = method))
     names(listScores) <- method
@@ -129,7 +129,7 @@ ROCAUC_Scores_Calculate <- function(data, metadata, gene_sets, method = c("logme
 #'
 ROC_Scores <- function(data, metadata, gene_sets, method = c("logmedian","ssGSEA","ranking","all"), variable,
                        colors = c(logmedian = "#3E5587", ssGSEA = "#B65285", ranking = "#B68C52"), grid = TRUE, spacing_annotation=0.3, ncol=NULL, nrow=NULL, mode=c("simple","medium","extensive"), widthTitle = 18, title=NULL, titlesize=12) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   data_ROCAUC <- ROCAUC_Scores_Calculate(data = data, metadata = metadata, gene_sets = gene_sets, method = method, variable = variable, mode = mode)
 
   plot_list <- list()
@@ -333,7 +333,7 @@ ROC_Scores <- function(data, metadata, gene_sets, method = c("logmedian","ssGSEA
 #'
 #' @export
 AUC_Scores <- function(data, metadata, gene_sets, method = c("logmedian", "ssGSEA", "ranking", "all"), mode = c("simple","medium","extensive"), variable, nrow = NULL, ncol = NULL, limits = NULL, widthTitle = 22, titlesize = 12, ColorValues = c("#F9F4AE", "#B44141"), title = NULL) {
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # Calculate the AUC scores using the given method
   auc_list <- ROCAUC_Scores_Calculate(data = data, metadata = metadata, gene_sets = gene_sets, method = method, variable = variable, mode = mode)
 
diff --git a/R/ROCandAUCplot.R b/R/ROCandAUCplot.R
@@ -108,7 +108,7 @@ ROCandAUCplot <- function(data, metadata,
                           roc_params = list(),     # ROC-specific parameters
                           auc_params= list(),
                           commomplot_params = list()) {  # Additional parameters for combined plots
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # If genes are not specified, use all available genes
   if (is.null(genes)) {
     genes <- rownames(data)
diff --git a/R/Score_VariableAssociation.R b/R/Score_VariableAssociation.R
@@ -152,6 +152,7 @@ Score_VariableAssociation <- function(data, metadata, cols, method=c("logmedian"
                                       continuous_color = "#8C6D03", color_palette = "Set2", printplt =TRUE){
   method <- match.arg(method)  # Validate method input
   mode <- match.arg(mode)
+  data <- as.data.frame(data) # Ensure data is a data frame
   # calculate scores for a given metric
   df_ranking <- CalculateScores(data = data, metadata = metadata, method = method, gene_sets = gene_set)[[1]] # if more than one gene set is provided, only results for the first one will be returned
 
diff --git a/R/VariableAssociation.R b/R/VariableAssociation.R
@@ -267,7 +267,7 @@ VariableAssociation <- function(method = c("ssGSEA", "logmedian", "ranking", "GS
                                 printplt = TRUE) {
   method <- match.arg(method)
   mode <- match.arg(mode)
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   if (method == "GSEA") {
     result <- GSEA_VariableAssociation(
       data = data,
diff --git a/R/VisualiseIndividualGenes.R b/R/VisualiseIndividualGenes.R
@@ -133,7 +133,7 @@
 #' @export
 VisualiseIndividualGenes <- function(type, data, genes, metadata = NULL, ...) {
   args <- list(...)
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   # Define required additional arguments for each plot type (besides data, genes, metadata)
   required_args <- list(
     violin      = c("GroupingVariable"),
diff --git a/R/calculateDE.R b/R/calculateDE.R
@@ -70,7 +70,7 @@
 #' @export
 calculateDE <- function(data, metadata=NULL, variables=NULL, modelmat = NULL, contrasts = NULL, ignore_NAs = FALSE) {
 
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   remove_prefix <- function(colnames_vector, prefixes) {
     for (prefix in prefixes) {
       new_colnames <- gsub(paste0("^", prefix), "", colnames_vector)
diff --git a/R/plotPCA.R b/R/plotPCA.R
@@ -70,7 +70,7 @@
 #' @importFrom ggpubr ggarrange
 #' @export
 plotPCA <- function(data, metadata=NULL, genes=NULL, scale=FALSE, center=TRUE, PCs=list(c(1,2)), ColorVariable=NULL,ColorValues=NULL,pointSize=5,legend_nrow=2, legend_position=c("bottom","top","right","left"),ncol=NULL, nrow=NULL){
-
+  data <- as.data.frame(data) # Ensure data is a data frame
   legend_position <- match.arg(legend_position)
 
   if (is.null(metadata) & !is.null(ColorVariable)) stop("ColorVariable only available when metadata is specified.")
