Merge pull request #15 from RibosomeCRCL/Dev_Allyson

Dev allyson

Author: AllysonMoureaux

R/get_diff_sites_summary.R

@@ -32,9 +32,10 @@ get_diff_sites_summary <- function(ribo = ribo, pthr = 0.05, condition_col = con
   qcdata <- ribo[[2]] 
 
   #----------------------------------------ANOVA test---------------------------
-  if(length(unique(qcdata$condition_col)) > 2){
+  if(length(unique(qcdata[[condition_col]])) > 2){
   compute_pval_anova <- rRMSAnalyzer::compute_pval(ribo = ribo, test = "anova", condition_col = condition_col) 
   a <-rRMSAnalyzer::plot_stat(ribo = ribo, compute_pval = compute_pval_anova, pthr = pthr, condition_col = condition_col, cscore_cutoff = cscore_cutoff)
+  sites_anova <- a[["plot_env"]][["significant_sites"]]
   }
 
   #----------------------------------------Welch test---------------------------
@@ -57,9 +58,10 @@ get_diff_sites_summary <- function(ribo = ribo, pthr = 0.05, condition_col = con
   }
 
   #----------------------------------------Kruskal test-------------------------
-  if(length(unique(qcdata$condition_col)) > 2){
+  if(length(unique(qcdata[[condition_col]])) > 2){
     compute_pval_kruskal <- rRMSAnalyzer::compute_pval(ribo = ribo, test = "kruskal", condition_col = condition_col)
     k <-rRMSAnalyzer::plot_stat(ribo = ribo, compute_pval = compute_pval_kruskal, pthr = pthr, condition_col = condition_col, cscore_cutoff = cscore_cutoff)
+    sites_kw <- k[["plot_env"]][["significant_sites"]]
   }
 
   #----------------------------------------Wilcoxon test------------------------
@@ -75,9 +77,9 @@ get_diff_sites_summary <- function(ribo = ribo, pthr = 0.05, condition_col = con
 
     ribo_filtered_x <- keep_ribo_samples(ribo_filtered_x,kept_samples_x)
 
-    compute_pval_wolcoxon <- rRMSAnalyzer::compute_pval(ribo = ribo_filtered_x, test = "wilcoxon", condition_col = condition_col)
+    compute_pval_wilcoxon <- rRMSAnalyzer::compute_pval(ribo = ribo_filtered_x, test = "wilcoxon", condition_col = condition_col)
 
-    x <- rRMSAnalyzer::plot_stat(ribo = ribo_filtered_x, compute_pval = compute_pval_wolcoxon, pthr = pthr, condition_col = condition_col, cscore_cutoff = cscore_cutoff)
+    x <- rRMSAnalyzer::plot_stat(ribo = ribo_filtered_x, compute_pval = compute_pval_wilcoxon, pthr = pthr, condition_col = condition_col, cscore_cutoff = cscore_cutoff)
     x_list[[i]] <- x 
   }
 
@@ -92,8 +94,8 @@ get_diff_sites_summary <- function(ribo = ribo, pthr = 0.05, condition_col = con
     comp_id <- paste0(case, "_vs_", ctrl)
 
     # Extract significant sites from tests
-    sites_welch <- w_list[[i]][["plot_env"]][["significant_sites"]]
-    sites_wilcox <- x_list[[i]][["plot_env"]][["significant_sites"]]
+    sites_welch <- unique(unlist(lapply(w_list, function(w) w[["plot_env"]][["significant_sites"]]))) 
+    sites_wilcox <- unique(unlist(lapply(x_list, function(x) x[["plot_env"]][["significant_sites"]])))
 
     # Add to the list
     comparison_results[[comp_id]] <- list(
@@ -106,15 +108,13 @@ get_diff_sites_summary <- function(ribo = ribo, pthr = 0.05, condition_col = con
   results <- list()
 
   # Identify all significant sites in at least on test
-  if(length(unique(qcdata$condition_col)) > 2){
+  if(length(unique(qcdata[[condition_col]])) > 2){
     sites_anova <- a[["plot_env"]][["significant_sites"]]
     sites_kw <- k[["plot_env"]][["significant_sites"]]
   }
-  sites_welch <- w_list[[i]][["plot_env"]][["significant_sites"]]
-  sites_wilcox <- x_list[[i]][["plot_env"]][["significant_sites"]]
 
   # all sites
-  if(length(unique(qcdata$condition_col)) > 2){
+  if(length(unique(qcdata[[condition_col]])) > 2){
   all_sites <- unique(c(sites_anova, sites_kw, sites_welch, sites_wilcox))
   } else {
     all_sites <- unique(c(sites_welch, sites_wilcox))
@@ -124,7 +124,7 @@ get_diff_sites_summary <- function(ribo = ribo, pthr = 0.05, condition_col = con
   results$Site <- all_sites
 
   # Add Anova et Kruskal column if groups > 2 
-  if (length(unique(qcdata$condition_col)) > 2) {
+  if (length(unique(qcdata[[condition_col]])) > 2) {
     results$Multiple_comparison_Anova <- ifelse(all_sites %in% sites_anova, "Significant", "NS")
     results$Multiple_comparison_Kruskal_Wallis <- ifelse(all_sites %in% sites_kw, "Significant", "NS")
     }

inst/rmd/diff_sites.Rmd

@@ -504,7 +504,7 @@ All resulting p-values were adjusted for multiple testing using the False Discov
 <br>
 \> Summary of Differentially Methylated rRNA 2′Ome Sites > (Only significant sites in at least one test are shown)
 
-```{r}
+```{r, warning=FALSE, message=FALSE}
 summary_table <- rRMSAnalyzer::get_diff_sites_summary(ribo = ribo_to_check, condition_col = params$condition_col, comparisons = params$comparisons)
 
 DT::datatable(

vignettes/Model_human.Rmd

@@ -154,7 +154,7 @@ ribo_adj_small <- keep_ribo_samples(ribo_adj,c("sample1","sample2","sample3","sa
 ### To remove a small number of samples  
 
 ```{r, eval=FALSE}
-ribo_adj_small <- remove_ribo_samples(ribo_adj,c("xxxx","xxxccc","RNA_ref-1","RNA_ref-2"))
+ribo_adj_small <- remove_ribo_samples(ribo_adj,c("xxxx","xxxccc","RNA_ref-1","RNA_ref-2")) # ribo_toy : ...,c("RNA1", "RNA2"))
 ```
 
 ### Uniformisation of the RiboClass name
@@ -211,10 +211,12 @@ To do only if you want to compare a subset of your samples (i.e., not all the sa
 
 ```{r, eval=FALSE}
 kept_samples <- ribo_adj_annot$metadata %>%
-  dplyr::filter(!is.na(variableA)) %>% # keep lines that are not "NA"
+  # keep lines that are not "NA"
+  dplyr::filter(!is.na(variableA)) %>% #ribo_toy (!is.na(comp1))
   dplyr::pull(samplename)
 
-ribo_adj_annot_variableA <- keep_ribo_samples(ribo_adj_annot,kept_samples) #create a new RiboClass including the subdata variableA
+#create a new RiboClass including the subdata variableA
+ribo_adj_annot_variableA <- keep_ribo_samples(ribo_adj_annot,kept_samples) #ribo_toy ribo_adj_annot_comp1 <- ...
 ```
 
 #### Create mandatory comparison table for the report_diff_sites 
@@ -231,15 +233,16 @@ comparisons <- tibble::tibble(
 
 ```{r, eval=FALSE}
 report_2ome_sites(ribo = ribo_adj_annot_variableA, specie = "human", condition_col = "variableA", project_name = "name", comments = "./path/to/comment_2ome_variableA.Rmd") 
-#condition_col: change the name of the column to perform new analysis
+#condition_col: change the name of the column to perform new analysis. ribo_toy condition_col = "comp1"
 #comments: change the path and name of your files with your comments if required
 ```
 
 #### Generate a report to perform site-by-site 2’Ome comparison
 
 ```{r, eval=FALSE}
 report_diff_sites(ribo = ribo_adj_annot_variableA, specie = "human", condition_col = "variableA", project_name = "name", comparisons = comparisons, comments = "./comment_diff_site_variableA.Rmd")
-#condition_col: change the name of the column to perform new analysis
+#ribo_toy : ribo = ribo_adj_annot_comp1
+#condition_col: change the name of the column to perform new analysis . ribo_toy condition_col = "comp1"
 #comparisons: name of the comparison table
 #comments: change the name of your files with your comments if required
 ```
@@ -251,7 +254,7 @@ Re-do all the steps of [Reports for comparisons explained in column 1] but using
 ## Export data
 
 ```{r, eval = FALSE}
-ribo_df <- extract_data(ribo_adj_annot,
+ribo_df <- extract_data(ribo_adj_annot, #change for the annotated RiboClass if needed
                         col = "cscore",
                         only_annotated = TRUE)