updated

TP_Suspect_List/runme_SuspectList.R

+## Script to Screen TPs
+## Emma Schymanski & Jessy Krier, 2 May 2020
+
+
+
+###### Directory and Package Setup #####
+
+##If you don't have the latest RChemMass, reinstall:
+library(devtools)
+install_github("schymane/RChemMass", dependencies = F)
+
+##If you have the latest RChemMass, do:
+library(RChemMass)
+
+##Set wd:
+base_dir <- "/Users/jessykrier/Dokumente/Uni.lu/Master in Systems Biology/4.Semester/MasterThesis/Master_Thesis/Compound_List_TransfProducts/"
+setwd(base_dir)
+
+##get the latest newest functions: 
+extr_anno_url <- "https://git-r3lab.uni.lu/eci/pubchem/-/raw/master/annotations/tps/extractAnnotations.R?inline=false"
+download.file(extr_anno_url,paste0(base_dir,"extractAnnotations.R"))
+source("extractAnnotations.R")
+
+## get HSDB file:
+hsdb_file <- paste0(base_dir,"Metabolism_Metabolites_HSDB_merged.csv")
+
+
+
+##### Start TP searching using Suspects file #####
+
+cmpd_info <- read.csv("TPSuspects_CompTox_wPCIDs.csv", stringsAsFactors = F)
+
+# #get PubChem CID and the parent CID based on InChIKey
+# # we will use the parent CID from now on ... 
+# for (i in 1:length(cmpd_info$INPUT)) {
+#   PCID <- getPcId(cmpd_info$INCHIKEY[i])
+#   cmpd_info$PCID[i] <- PCID
+#   cmpd_info$Parent_CID[i] <- getPCIDs.CIDtype(PCID,type="parent")
+# }
+# write.csv(cmpd_info, "Jessys_Suspects_CompTox_wPCIDs.csv", row.names = F)
+
+
+
+##### For the first run #####
+#i <- 73
+
+for (i in 1:nrow(cmpd_info)) {
+  
+
+  query_CID <- as.numeric(cmpd_info$Parent_CID[i])
+  # this gets you a download file and unique CIDs of the transformations section
+  trans_out <- getPcCand.trans(query_CID)
+  trans_cid_file <- trans_out$TP_File_Name
+  trans_tp_cids <- trans_out$TP_Unique_CIDs
+  # this gets you a list of unique CIDs from HSDB, plus a download file
+  hsdb_out <- getPcAnno.TPcids(query_CID, hsdb_file)
+  hsdb_cid_file <- hsdb_out$TP_File_Name
+  hsdb_tp_cids <- hsdb_out$TP_Unique_CIDs
+  
+ # }, 
+  
+
+}
+
+
+
+#### For the Shiny files ####
+
+MoNA_pest_list <- read.csv("Match_MoNA_VeryGood.csv", stringsAsFactors = F)
+
+#k <- 3
+
+
+for (k in 1:nrow(MoNA_pest_list)) {
+  
+
+    query_CID <- as.numeric(MoNA_pest_list$CID[k])
+    # this gets you a download file and unique CIDs of the transformations section
+    trans_out <- getPcCand.trans(query_CID)
+    trans_cid_file <- trans_out$TP_File_Name
+    trans_tp_cids <- trans_out$TP_Unique_CIDs
+    # this gets you a list of unique CIDs from HSDB, plus a download file
+    hsdb_out <- getPcAnno.TPcids(query_CID, hsdb_file)
+    hsdb_cid_file <- hsdb_out$TP_File_Name
+    hsdb_tp_cids <- hsdb_out$TP_Unique_CIDs
+    
+    file_name_1 <- paste("CID_",MoNA_pest_list$CID[k], "_norman_s60.csv",  sep = "")
+    file_name_2 <- paste("Metabolism_Metabolites_HSDB_merged_",MoNA_pest_list$CID[k], "_selected.csv",  sep = "")
+    print(paste(file.exists(file_name_1),file_name_1))
+    print(paste(file.exists(file_name_2),file_name_2))
+    
+       if(file.exists(file_name_1) | file.exists(file_name_2)){
+        
+         if(!file.exists(file_name_2)){
+           merged_tp_cids <- unique(c(query_CID,trans_tp_cids))
+         }
+         else{
+          #you might want to create a new file name, if so, enter here:
+          hsdb_selected_file <- sub(".csv","_selected.csv",hsdb_cid_file)
+          hsdb_info <- read.csv(hsdb_selected_file, stringsAsFactors = F)
+  
+          ##select only those CIDs to use, collapse and uniquify
+          hsdb_tp_cids <- hsdb_info$selected_tp_cids[which(hsdb_info$use==TRUE)]
+          hsdb_tp_cids <- paste(hsdb_tp_cids,collapse="|")
+          hsdb_tp_cids <- suppressWarnings(as.numeric(unique(strsplit(hsdb_tp_cids,"|",fixed=T)[[1]])))
+  
+          # this gets you all CIDs from both sources, including parent
+          merged_tp_cids <- unique(c(query_CID,trans_tp_cids,hsdb_tp_cids))
+          
+          }
+          ##### Create files for Shinyscreen #####
+          ## note this first file can run a while
+          MSInfo_file <- getMSInfo.cids(merged_tp_cids,paste0(query_CID,"_MSInfo.csv")) #getting all mass info
+          getMSInfo.files(MSInfo_file, set_id = query_CID)
+         
+
+     }
+    
+
+
+}
+
+
+
+
+##### Create files for Zenodo/PubChem #####
+
+## we also now need to create files to save the new HSDB entries in the transformations format
+hsdb_desc <- paste0("HSDB is a toxicology database that focuses on the ",
+                    "toxicology of potentially hazardous chemicals. See ",
+                    "https://pubchem.ncbi.nlm.nih.gov/source/11933")
+
+dataset_DOI <- "10.5281/zenodo.3827487"
+
+dataset_desc <- "Metabolites extracted from HSDB and validated by ECI/PubChem"
+
+# for parent compounds (atrazine)
+trans_file <- createFiles.trans(hsdb_tp_cids, hsdb_selected_file, MSInfo_file, 
+                                hsdb_desc, dataset_DOI, dataset_desc, 
+                                file_name = "")
+
+# for metabolites (deethylatrazine)
+trans_file <- createFiles.trans(hsdb_tp_cids, hsdb_selected_file, MSInfo_file, 
+                                hsdb_desc, dataset_DOI, dataset_desc, 
+                                file_name = "",startCIDisPredecessor = F)
+
+
+