First steps normalisation and correlation checks. Filtered out samples that...

First steps normalisation and correlation checks. Filtered out samples that did not pass quality checks.

Orsay_transcriptomics.R

@@ -4,33 +4,37 @@
 
 
 ###Set your work directory
-setwd("H:/Nemawork/Projects/Orsay_transcriptomics")
-workwd <- getwd()
-
-Lisa <- FALSE
-if(Lisa){
-###Load pre-made functions
-    #uses 
-    #     transcriptomics functions https://git.wur.nl/mark_sterken/Transcriptomics.workbench
-git_dir <- "H:/Nemawork/Projects_R_zone/Git"
-source(paste(git_dir,"/Loader_git.R",sep=""))
-
-###Set data locations
-resources_dir <- paste(git_dir,"/../R_Resources/",sep="")
-support_dir <- paste(git_dir,"/Orsay_transcriptomics/")
-}
-
-if(Lisa==FALSE){
-    ###Load pre-made functions
-        #uses 
-        #     transcriptomics functions https://git.wur.nl/mark_sterken/Transcriptomics.workbench
-    git_dir <- "H:/Nemawork/Projects_R_zone/Git"
-    source(paste(git_dir,"/Loader_git.R",sep=""))
+    setwd("H:/Projects/OrV_transcriptomics/Analysis")
+    workwd <- getwd()
     
-    ###Set data locations
-    resources_dir <- paste(git_dir,"/../R_Resources/",sep="")
-    support_dir <- paste(git_dir,"/Orsay_transcriptomics/Support_files",sep="")
-}
+    Lisa <- TRUE
+    
+        if(Lisa){
+        ###Load pre-made functions
+            #uses 
+            #     transcriptomics functions https://git.wur.nl/mark_sterken/Transcriptomics.workbench
+        #git_dir <- "H:/R/Git"
+        C.ele.drive <- "H:/R/R_zone"
+        Organism <- "ele"
+        source(paste("H:/R/R_zone/R_functions/Loader_v3.r",sep=""))
+    
+        ###Set data locations
+        resources_dir <- "H:/R/R_zone/R_Resources/"
+        support_dir <- "H:/Projects/OrV_transcriptomics/Raw"
+        
+    }
+    
+    if(Lisa==FALSE){
+        ###Load pre-made functions
+            #uses 
+            #     transcriptomics functions https://git.wur.nl/mark_sterken/Transcriptomics.workbench
+        git_dir <- "H:/Nemawork/Projects_R_zone/Git"
+        source(paste(git_dir,"/Loader_git.R",sep=""))
+        
+        ###Set data locations
+        resources_dir <- paste(git_dir,"/../R_Resources/",sep="")
+        support_dir <- paste(git_dir,"/Orsay_transcriptomics/Support_files",sep="")
+    }
 
 
 ################################################################################
@@ -60,7 +64,8 @@ if(install){
     library(openxlsx)
     library("rmarkdown")
     library(tidyverse)
-
+    library("ggplot2")
+    library("dplyr")
 
 ################################################################################
 ###Plotting theme, colours
@@ -103,21 +108,28 @@ if(install){
     colScale <- scale_colour_manual(name = "Treatment",values = myColors)
     fillScale <- scale_fill_manual(name = "Treatment",values = myColors)
     
+    cols <- c("CB4856" = "#0072B2","N2"="#E69F00", "JU1580" = "#00CC33")
+    
 
 ################################################################################
 ###Load targets, normalize & quality check
 ################################################################################
- 
+    
     targets <- read.delim(paste(support_dir,"/Target_OrV.txt",sep="")); head(targets)
 
-    rg.norm <- agilent.limma.norm(targets=targets,array_dir=paste(workwd,"/../../Projects_data_reagents/MA_elegans/",sep=""),save_dir=paste(workwd,"/Normalized/",sep=""),filename="OrV")
-
+    #Removed 4 samples (2 arrays) with low quality before normalisation: sample number 539, 665, 554 and 607. 
+    #Other samples that also have a low quality, but are not on the same array : 596, 728, 451, 470. They are manually filtered. 
+    #Also the samples with low correlations in the heatmap: 727 and 474 are removed manually.
+    rg.norm <- agilent.limma.norm(targets=targets,array_dir=paste(workwd,"/../Raw/",sep=""),save_dir=paste(workwd,"/Normalized/",sep=""),filename="OrV", bad.arrays = c(20, 10, 31, 11))
+    rg.norm <- rg.norm[,-c(1, 9, 10, 26, 27, 28)]
+    
     trans.int <- transform.norm(rg.norm,save_dir=paste(workwd,"/Normalized/",sep=""),filename="OrV"); lapply(trans.int,head)
     
     ###Check quality
-    correlsums <- correlation.check(trans.int,filename="OrV",save_dir=paste(workwd,"/Normalized/",sep=""))
+
+    correlsums <- correlation.check(trans.int,filename="OrV",save_dir=paste(workwd,"/QC/",sep=""))
     
-    genes.check(trans.int,spot.id=Agilent.Ce.V2$gene_public_name,genes.to.check = c("gst-27","vit-1","vit-2","hsp-16.2","hsp-16.41","F26F2.1"),filename="OrV",save_dir=paste(workwd,"/Normalized/",sep=""))
+    genes.check(trans.int,spot.id=agi.id$Public_name,genes.to.check = c("gst-27","vit-1","vit-2","hsp-16.2","hsp-16.41","F26F2.1"),filename="OrV",save_dir=paste(workwd,"/GC/",sep=""))
 
     
     colnames.sep <- ":"