Merge branch 'update_workflows' into 'main'

restructure workflows for reproducibility

See merge request brilator/samplearc_rnaseq!2

README.md

@@ -91,4 +91,12 @@ DB_161 \
 DB_163 \
 DB_165
 cd $arc_root
-```
\ No newline at end of file
+```
+
+## Make workflows a bit more representative and reproducible (24.03.2022)
+
+## generate a common arc_root pointer
+
+```bash
+echo "~/03DataPLANT_gitlab/samplearc_rnaseq/" > workflows/_arc_local_wd
+```

workflows/.Rhistory

-ARC_root="~/03_DataPLANT_gitlab/samplearc_rnaseq/"

workflows/01_KallistoQuant.sh

@@ -3,11 +3,12 @@
 #### To be replaced by CWL routine
 ########################
 
-ARC_root=~/samplearc_rnaseq/
-cd $ARC_root'workflows/'
-
+# Execute within <ARC root>/workflows
 # chmod a+x 01_KallistoQuant.sh
 # ./01_KallistoQuant.sh > $ARC_root'runs/01_kallisto.log' 2>&1 &
+# pointers to and from `runs` need to be replaced
+
+ARC_root=$(cat ./_arc_local_wd)
 
 ########################
 
@@ -21,7 +22,7 @@ kallisto cite
 
 ### Build index 
 
-kall_ref=$ARC_root'externals/Talinum.gm.CDS.nt.fa'
+kall_ref=$ARC_root'studies/TalinumGenomeDraft/resources/Talinum.gm.CDS.nt.fa'
 kallisto index -i $ARC_root'runs/01_kallisto_index' $kall_ref
 
 ### Align reads
@@ -32,11 +33,11 @@ mkdir $ARC_root'/runs/01_kallisto_results/'
 
 for j in $ILLUMINASAMPLES; do
 
-		sampleName=$(echo $j | sed -e 's|.*/||' | cut -c -6)  # cut away path. retain only first six chars of file name
-		echo $sampleName
-		
-		kallisto quant --single -b 100 -t 30 -l 200 -s 20 -i $ARC_root'/runs/01_kallisto_index' -o $ARC_root'/runs/01_kallisto_results/'$sampleName $j
+	sampleName=$(echo $j | sed -e 's|.*/||' | cut -c -6)  # cut away path. retain only first six chars of file name
+	echo $sampleName
+	
+	kallisto quant --single -b 100 -t 30 -l 200 -s 20 -i $ARC_root'/runs/01_kallisto_index' -o $ARC_root'/runs/01_kallisto_results/'$sampleName $j
 
-	    echo 'Kallisto done'
+	echo 'Kallisto done'
 
 done
\ No newline at end of file

workflows/03_KallistoCollect.R

@@ -4,8 +4,12 @@
 #### To be replaced by CWL routine
 ########################
 
-ARC_root="~/Hackathon_ARCexample_rnaseq/"
-setwd(paste0(ARC_root, 'workflows/'))
+# Execute within <ARC root>/workflows
+# Rscript 03_KallistoCollect.R
+# pointers to and from `runs` need to be replaced
+
+ARC_root=readLines("./_arc_local_wd")
+
 
 ########################
 
@@ -29,23 +33,29 @@ library(openxlsx)
 
 ## read experimental metadata from isa.assay wb
 
-isa_assay <- paste0(ARC_root, 'assays/Talinum_RNASeq_minimal/assay.isa.xlsx')
+isa_assay <- paste0(ARC_root, 'assays/Talinum_RNASeq_minimal/isa.assay.xlsx')
+isa_study <- paste0(ARC_root, 'studies/TalinumFacultativeCAM/isa.study.xlsx')
 
-assay_data <- merge(readWorkbook(isa_assay, "1SPL01_plants", startRow = 2), 
-                    readWorkbook(isa_assay, "3ASY01_RNASeq", startRow = 2), 
-                    by = "Sample.Name"
+study_data <- readWorkbook(isa_study, "plant_growth", startRow = 1)
+
+assay_data <- merge(readWorkbook(isa_assay, "2EXT01_RNA", startRow = 1), 
+                    readWorkbook(isa_assay, "3ASY01_RNASeq", startRow = 1), 
+                    by.x = "Sample.Name", 
+                    by.y = "Source.Name"
                     )
 
+assay_data <- merge(study_data, assay_data,  by.x = "Sample.Name", by.y = "Source.Name")
+
 ## remove empty cols
 assay_data <- assay_data[, !apply(assay_data, 2, function(x){sum(is.na(x)) == nrow(assay_data)})]  
 
 # Pointer to kallisto results folder
-base_dir <- paste0(ARC_root, '/runs/01_kallisto_results/')
+base_dir <- paste0(ARC_root, 'runs/01_kallisto_results/')
 
 # A list of paths to the kallisto results indexed by the sample IDs is collated with
 kal_dirs <- dir(base_dir, full.names = T) ## Sleuth requires full paths
 
-s2c <- assay_data[order(assay_data$Sample.Name), c('Sample.Name', "Characteristics.[Photosynthesis.mode]")]
+s2c <- assay_data[order(assay_data$Sample.Name), c('Sample.Name', "Factor.[Photosynthesis.mode]")]
 # For kallisto / sleuth: 's2c' (sample_to_covariates) must contain a column named 'sample'
 colnames(s2c) <- c("sample", "Photosynthesis.mode")
 

workflows/04_Sleuth.R

@@ -4,8 +4,11 @@
 #### To be replaced by CWL routine
 ########################
 
-ARC_root="~/Hackathon_ARCexample_rnaseq/"
-setwd(paste0(ARC_root, 'workflows/'))
+# Execute within <ARC root>/workflows
+# Rscript 04_Sleuth.R
+# pointers to and from `runs` need to be replaced
+
+ARC_root=readLines("./_arc_local_wd")
 
 ########################
 

workflows/05_plot_shinyPrep.R

@@ -4,8 +4,11 @@
 #### To be replaced by CWL routine
 ########################
 
-ARC_root="~/Hackathon_ARCexample_rnaseq/"
-setwd(paste0(ARC_root, 'workflows/'))
+# Execute within <ARC root>/workflows
+# Rscript 05_plot_shinyPrep.R
+# pointers to and from `runs` need to be replaced
+
+ARC_root=readLines("./_arc_local_wd")
 
 ########################
 

workflows/_arc_local_wd

+~/03DataPLANT_gitlab/samplearc_rnaseq/