updated mutant matrix file and run_GSEA.R

R/run_GSEA.R

@@ -15,7 +15,7 @@ kegg_to_t2g = function(data_set){
   GSEA_kegg_list <- list()
   for (org_id in kegg_orgids){
     tmp_subset <- data_set[[org_id]]
-    tmp_subset$label_joined <- paste(tmp_subset$pathway_id, tmp_subset$pathway_name, sep="_")
+    tmp_subset$label_joined <- paste(gsub("path:","",tmp_subset$pathway_id), gsub(" - Mycobacterium tuberculosis CDC1551| - Mycobacterium tuberculosis H37Rv","",tmp_subset$pathway_name), sep="_")
     GSEA_kegg_list[[org_id]] <- tmp_subset[,c("label_joined","uniprot_id")]
   }
   return(GSEA_kegg_list)
@@ -44,31 +44,32 @@ kegg_to_t2g = function(data_set){
 
 
 run_GSEA = function(data_set, kingdom, pvalCOff,MAIN_PATH, kegg_list,mycobrowser_data){
+  set.seed(1234)
   if(!(kingdom %in% c("GO_CC","GO_BP","GO_MF", "KEGG", "MYCOBROWSER_FC", "MYCOBROWSER_FU", "MYCOBROWSER_PR"))){print("WARNING, choose kingdom you want to analyze. Options:\"GO_CC\",\"GO_BP\",\"GO_MF\", \"KEGG\", \"MYCOBROWSER_FC\", \"MYCOBROWSER_FU\", \"MYCOBROWSER_PR\".")}
   if(kingdom == "GO_CC"){
     print("You are running GSEA on GeneOntology category \"Cellular Compartment\".")
     load(paste(MAIN_PATH, "annotation_files", "cc_table_for_gmt.Rdata", sep="/")) # loads the data set: cc_table_for_gmt
-    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=cc_table_for_gmt, by="fgsea")) #cutoff for adjusted pvalue
+    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=cc_table_for_gmt, by="fgsea", seed=T)) #cutoff for adjusted pvalue
   }
   if(kingdom == "GO_BP"){
     print("You are running GSEA on GeneOntology category \"Biological Process\".")
     load(paste(MAIN_PATH, "annotation_files", "bp_table_for_gmt.Rdata", sep="/")) # loads the data set: bp_table_for_gmt
-    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=bp_table_for_gmt, by="fgsea"))
+    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=bp_table_for_gmt, by="fgsea", seed=T))
   }
   if(kingdom == "GO_MF"){
     print("You are running GSEA on GeneOntology category \"Molecular Function\".")
     load(paste(MAIN_PATH, "annotation_files", "mf_table_for_gmt.Rdata", sep="/")) # loads the data set: mf_table_for_gmt
-    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mf_table_for_gmt, by="fgsea"))
+    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mf_table_for_gmt, by="fgsea", seed=T))
   }
   if(kingdom == "KEGG"){
-    print("You are running GSEA on kegg pathways. The code runs for every Mtb organism ID on KEGG separately, so the output will be a list, not a table as when running the function for GO terms.")
+    print("You are running GSEA on kegg pathways. The code runs for every Mtb organism ID on KEGG separately, so the output will be a list, not a table as when running the function for GO terms. Some KEGG organism IDs have contain different protein identifiers, not recognized by the GSEA Function. This could be corrected, but because they are highly redundant, this is not done here. Most important are results from organism IDs: mtu & mtv, which both are Mtb H37Rv.")
     GSEA_kegg_list = kegg_to_t2g(kegg_list)
     GSEA_kegg_result_list <- list()
     kegg_orgids <- names(GSEA_kegg_list)
     for (org_id in kegg_orgids){
       message(paste("Running GSEA on comparison_",org_id," right now. Computing..."))
       runfun <- tryCatch(
-        res_GSEA <- GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=GSEA_kegg_list[[org_id]], by="fgsea"),
+        res_GSEA <- GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=GSEA_kegg_list[[org_id]], by="fgsea", seed=T),
         error = function(e) {
           message(paste("Running GSEA on the comparison: ",org_id," ended up in error...And Now for Something Completely Different."))
           NA
@@ -81,17 +82,17 @@ run_GSEA = function(data_set, kingdom, pvalCOff,MAIN_PATH, kegg_list,mycobrowser
   if(kingdom == "MYCOBROWSER_FC"){
     print("You are running GSEA on mycobrowser category \"Functional Category\".")
     mycobrowser_FunCat   <- unique(mycobrowser_data[,c("Functional_Category","UniProt_AC")])
-    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mycobrowser_FunCat, by="fgsea"))
+    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mycobrowser_FunCat, by="fgsea", seed=T))
   }
   if(kingdom == "MYCOBROWSER_FU"){
     print("You are running GSEA on mycobrowser category \"Function\".")
     mycobrowser_Function <- unique(mycobrowser_data[,c("Function","UniProt_AC")])
-    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mycobrowser_Function, by="fgsea"))
+    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mycobrowser_Function, by="fgsea", seed=T))
   }
   if(kingdom == "MYCOBROWSER_PR"){
     print("You are running GSEA on mycobrowser category \"Product\".")
     mycobrowser_Product  <- unique(mycobrowser_data[,c("Product","UniProt_AC")])
-    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mycobrowser_Product, by="fgsea"))
+    return(GSEA(rev(data_set), minGSSize = 1, maxGSSize = (Inf), eps = 0, pvalueCutoff = pvalCOff, pAdjustMethod = "BH", TERM2GENE=mycobrowser_Product, by="fgsea", seed=T))
   }
 }
 

R/run_QC.R

@@ -25,6 +25,7 @@ run_QC = function(input_data,qc_level){
     input_data$norm_quantity <- 2^(input_data$normalised_intensity_log2)
     # 4.3 CVs
     print(qc_cvs(data = input_data, grouping = pep_grouping_key, condition = r_condition, intensity = pep_quantity, plot = TRUE, plot_style = "violin"))
+    print("Plot below shows CVs post normalization by 'pep_quantity'.")
     print(qc_cvs(data = input_data, grouping = pep_grouping_key, condition = r_condition, intensity = norm_quantity, plot = TRUE, plot_style = "violin"))
     # 4.4 PCA qc_pca( data = input_data, sample = r_file_name, grouping = pep_grouping_key, intensity = normalised_intensity_log2, condition = r_condition, digestion = NULL, plot_style = "scree" )
     print(qc_pca(data = input_data, sample = r_file_name, grouping = pep_grouping_key, intensity = normalised_intensity_log2, condition = r_condition, components= c("PC1","PC2"), plot_style = "pca" ) )
@@ -35,6 +36,7 @@ run_QC = function(input_data,qc_level){
     print(qc_ids(data = input_data, sample = r_file_name, grouping = pep_grouping_key, intensity = norm_quantity, condition = r_condition, title = "Peptide identifications per sample", plot = TRUE ) )
     # 4.7 intensity distribution
     print(qc_intensity_distribution(data = input_data, sample = r_file_name, grouping = pep_grouping_key, intensity_log2 = transformed_quantity_log2 , plot_style = "boxplot") )
+    print("Plot below shows intensity distribution post normalization by 'pep_quantity'.")
     print(qc_intensity_distribution(data = input_data, sample = r_file_name, grouping = pep_grouping_key, intensity_log2 = normalised_intensity_log2, plot_style = "boxplot"))
     # 4.8 data completeness
     print(qc_data_completeness(data = input_data, sample = r_file_name, grouping = pep_grouping_key, intensity = normalised_intensity_log2, plot = TRUE) )
@@ -48,6 +50,7 @@ run_QC = function(input_data,qc_level){
     input_data$norm_quantity <- 2^(input_data$normalised_intensity_log2)
     # 4.3 CVs
     print(qc_cvs(data = input_data, grouping = pg_protein_groups, condition = r_condition, intensity = pg_quantity, plot = TRUE, plot_style = "violin"))
+    print("Plot below shows CVs post normalization by 'pg_quantity'.")
     print(qc_cvs(data = input_data, grouping = pg_protein_groups, condition = r_condition, intensity = norm_quantity, plot = TRUE, plot_style = "violin"))
     # 4.4 PCA qc_pca( data = input_data, sample = r_file_name, grouping = pg_protein_groups, intensity = normalised_intensity_log2, condition = r_condition, digestion = NULL, plot_style = "scree" )
     print(qc_pca(data = input_data, sample = r_file_name, grouping = pg_protein_groups, intensity = normalised_intensity_log2, condition = r_condition, components= c("PC1","PC2"), plot_style = "pca" ) )
@@ -58,6 +61,7 @@ run_QC = function(input_data,qc_level){
     print(qc_ids(data = input_data, sample = r_file_name, grouping = pg_protein_groups, intensity = norm_quantity, condition = r_condition, title = "Peptide identifications per sample", plot = TRUE ) )
     # 4.7 intensity distribution
     print(qc_intensity_distribution(data = input_data, sample = r_file_name, grouping = pg_protein_groups, intensity_log2 = transformed_quantity_log2 , plot_style = "boxplot") )
+    print("Plot below shows intensity distribution post normalization by 'pg_quantity'.")
     print(qc_intensity_distribution(data = input_data, sample = r_file_name, grouping = pg_protein_groups, intensity_log2 = normalised_intensity_log2, plot_style = "boxplot") )
     # 4.8 data completeness
     print(qc_data_completeness(data = input_data, sample = r_file_name, grouping = pg_protein_groups, intensity = normalised_intensity_log2, plot = TRUE) )

inst/annotation_files/PROTEOME_PIPELINE_MUTANTMATRIX.txt

@@ -9,8 +9,8 @@ N2030,N0157,L1,HF,L1_HF,rpoB,S450L,na,1
 N2031,N0157,L1,LF,L1_LF,rpoB,H445R,na,1
 N1888,N0145,L2,HF,L2_HF,rpoB,S450L,na,1
 N1890,N0145,L2,LF,L2_LF,rpoB,H445R,na,1
-N2502,N1283,L4,HF,L4_HF,rpoB,S450L,na,1
-N2850,N1283,L4,LF,L4_LF,rpoB,H445R,na,1
+N2502,N1283,L4,LF,L4_LF,rpoB,H445R,na,1
+N2850,N1283,L4,HF,L4_HF,rpoB,S450L,na,1
 N0157,N0157,L1,WT,L1_WT,WT,WT,1,0
 N0145,N0145,L2,WT,L2_WT,WT,WT,1,0
-N1283,N1283,L4,WT,L4_WT,WT,WT,1,0
\ No newline at end of file
+N1283,N1283,L4,WT,L4_WT,WT,WT,1,0