Add sample list file as argument

README.md

@@ -20,13 +20,14 @@ For now, scripts have to be launched from the clone directory.
    - `wrapper_gangstr.sh`
    - `wrapper_transfer.sh`
    - `wrapper_tredparse.sh`
-4. Optional: launch `launch_pipeline_ehdn_outlier.sh`: `nohup ./launch_pipeline_ehdn_outlier.sh &`. Dependencies:
+4. Optional: launch `launch_pipeline_ehdn_outlier.sh`: `nohup ./launch_pipeline_ehdn_outlier.sh samples.list &`. Dependencies:
    - `config.sh`
+   - `samples.list`
    - `pipeline_ehdn_outlier.sh`
    - `wrapper_ehdn_outlier.sh`
 5. Launch `launch_results.sh`: `nohup ./launch_results.sh samples.list &`. Dependencies:
    - `config.sh`
-   - `sample.list`
+   - `samples.list`
    - `patho.csv`
    - `getResults.py`
    - `launch_str_outliers.sh`

launch_pipeline.sh

@@ -6,7 +6,7 @@
 ## Author: Anne-Sophie Denommé-Pichon
 ## Description: script to launch the pipeline for STR detection. Receive multiple samples: one sample per line
 
-# $1 : first argument in the command line : a list containing one sample per line, for exemple samples.list
+# $1 : first argument in the command line : a list containing one sample per line, for example samples.list
 SAMPLES="$1"
 
 # Check if sample is specified

launch_pipeline_ehdn_outlier.sh

@@ -6,14 +6,23 @@
 ## Author : Anne-Sophie Denommé-Pichon
 ## Description : script to launch the pipeline for STR detection with EHDN outlier
 
+# $1 : first argument in the command line : a list containing one sample per line, for example samples.list
+SAMPLES="$1"
+
+# Check if samples list file is specified
+if [ -z "$SAMPLES" ]
+then
+    echo "List of samples is not specified"
+    echo "$(date +"%F_%H-%M-%S"): END"
+    exit 1
+fi
+
 # Source configuration file
 . "$(dirname "$0")/config.sh"
 
-(
-    cd "$OUTPUTDIR"
-    printf "%s\n" *
-) |
-    "$PARALLEL" \
-	--jobs "$PARALLEL_JOB_COUNT" \
-        --line-buffer \
-        "$(dirname "$0")/pipeline_ehdn_outlier.sh"
+# Parallel allow to parallelize the processing of multiple samples ({}: one line in samples.list ; $SAMPLES: the file name: samples.list)
+"$PARALLEL" \
+    --jobs "$PARALLEL_JOB_COUNT" \
+    --line-buffer \
+    "$(dirname "$0")/pipeline_ehdn_outlier.sh" {} "$SAMPLES" \
+    < "$SAMPLES"

launch_results.sh

@@ -6,7 +6,7 @@
 ## Author: Anne-Sophie Denommé-Pichon
 ## Description: script to launch the pipeline for getting STR detection results. Receive multiple samples: one sample per line
 
-# $1 : first argument in the command line : a list containing one sample per line, for exemple samples.list
+# $1 : first argument in the command line : a list containing one sample per line, for example samples.list
 SAMPLES="$1"
 
 # Check if sample is specified

launch_str_outliers.sh

@@ -9,7 +9,7 @@
 # Source configuration file
 . "$(dirname "$0")/config.sh"
 
-# $1 : first argument in the command line : a list containing one sample per line, for exemple samples.list
+# $1 : first argument in the command line : a list containing one sample per line, for example samples.list
 SAMPLES="$1"
 
 # Check if sample is specified

pipeline.sh

@@ -61,7 +61,7 @@ qsub -wd "$WD" -pe smp 4 -o "$LOGDIR" -e "$LOGDIR" -q "$COMPUTE_QUEUE" -N "gangs
 
 # Launch ehdn profile
 mkdir -p "$STRDIR/ehdn"
-qsub -wd "$WD" -pe smp 4 -o "$LOGDIR" -e "$LOGDIR" -q "$COMPUTE_QUEUE" -N "ehdn_$SAMPLE" -hold_jid "$TRANSFER_JOB" -v INPUTFILE="$INPUTFILE",OUTPUTPREFIX="$STRDIR/ehdn/$SAMPLE",LOGFILE="$LOGDIR/ehdn_$SAMPLE.$DATE.log" "$WD/wrapper_ehdn_profile.sh"
+qsub -wd "$WD" -pe smp 4 -o "$LOGDIR" -e "$LOGDIR" -q "$COMPUTE_QUEUE" -N "ehdn_$SAMPLE" -hold_jid "$TRANSFER_JOB" -v INPUTFILE="$INPUTFILE",OUTPUTPREFIX="$STRDIR/ehdn/$SAMPLE",LOGFILE="$LOGDIR/ehdn_profile_$SAMPLE.$DATE.log" "$WD/wrapper_ehdn_profile.sh"
 
 # Delete transfered bam and bai
 #if [ "x$TRANSFER" = "xyes" ]

pipeline_ehdn_outlier.sh

@@ -10,6 +10,15 @@
 . "$(dirname "$0")/config.sh"
 
 CASE="$1"
+SAMPLES="$2"
+DATE="$(date +"%F_%H-%M-%S")"
+WD="$(dirname "$(readlink -f "$0")")"
 
-qsub -wd "$(dirname "$(readlink -f "$0")")" -pe smp 1 -q "$COMPUTE_QUEUE" -N "ehdn_outlier_$CASE" -sync y -v CASE="$CASE" wrapper_ehdn_outlier.sh
+LOGDIR="$OUTPUTDIR/logs"
+STRDIR="$OUTPUTDIR/str"
+
+mkdir -p "$LOGDIR" "$STRDIR"
+
+
+qsub -wd "$WD" -pe smp 1 -o "$LOGDIR" -e "$LOGDIR" -q "$COMPUTE_QUEUE" -N "ehdn_outlier_$CASE" -sync y -v CASE="$CASE",SAMPLES="$SAMPLES",LOGFILE="$LOGDIR/ehdn_outlier_$CASE.$DATE.log" "$WD/wrapper_ehdn_outlier.sh"
 

str_outliers.py

@@ -102,7 +102,7 @@ def display_outliers(locus, limits, samples):
             print('Input file is empty', file=sys.stderr)
             sys.exit(1)
 
-    # outlier threshold (exemple: 5%)
+    # outlier threshold (example: 5%)
     for tool, tool_values in tools_values.items():
         # Test if there is at least one value given by the tool
         if tool_values:

wrapper_ehdn_outlier.sh

@@ -9,8 +9,13 @@
 # Source the configuration file
 . ./config.sh
 
+# Log file path option
+if [ -z "$LOGFILE" ]
+then
+    LOGFILE=ehdn_outlier.$(date +"%F_%H-%M-%S").log
+fi
+
 # Logging
-LOGFILE="$OUTPUTDIR/$CASE/ehdn/$CASE.ehdn_outlier.$(date +"%F_%H-%M-%S").log"
 exec 1>> "$LOGFILE" 2>&1
 echo "$(date +"%F_%H-%M-%S"): START"
 
@@ -22,21 +27,28 @@ then
     exit 1
 fi
 
+# Check if samples list file is specified
+if [ -z "$SAMPLES" ]
+then
+    echo "Sample list file is not specified"
+    echo "$(date +"%F_%H-%M-%S"): END"
+    exit 1
+fi
+
 # Generate manifest for one patient with all samples (to write lines in the file)
-cd "$OUTPUTDIR"
-for sample in *
+while read sample
 do
     # Check if str_profile.json exists
-    if [ -f "$OUTPUTDIR/$sample/ehdn/$sample.str_profile.json" ]
+    if [ -f "$OUTPUTDIR/$sample/str/ehdn/$sample.str_profile.json" ]
     then
         if [ "x$sample" = "x$CASE" ]
         then
-        echo -e "$sample\tcase\t$OUTPUTDIR/$sample/ehdn/$sample.str_profile.json"
+            echo -e "$sample\tcase\t$OUTPUTDIR/$sample/str/ehdn/$sample.str_profile.json"
         else
-        echo -e "$sample\tcontrol\t$OUTPUTDIR/$sample/ehdn/$sample.str_profile.json"
+            echo -e "$sample\tcontrol\t$OUTPUTDIR/$sample/str/ehdn/$sample.str_profile.json"
         fi
     fi
-done > "$OUTPUTDIR/$CASE/ehdn/$CASE.manifest.tsv"
+done < "$SAMPLES" > "$OUTPUTDIR/$CASE/str/ehdn/$CASE.manifest.tsv"
 
 ehdn_outlier_exitcode=$?
 
@@ -50,8 +62,8 @@ fi
 # Generate multisampleprofile for one patient with all samples
 "$EHDN" merge \
     --reference "$REFERENCE" \
-    --manifest "$OUTPUTDIR/$CASE/ehdn/$CASE.manifest.tsv" \
-    --output-prefix "$CASE/ehdn/$CASE"
+    --manifest "$OUTPUTDIR/$CASE/str/ehdn/$CASE.manifest.tsv" \
+    --output-prefix "$CASE/str/ehdn/$CASE"
 
 ehdn_outlier_exitcode=$?
 
@@ -64,9 +76,9 @@ fi
 
 # Run locus-based comparison analysis
 "$EHDN_OUTLIER" locus \
-    --manifest "$OUTPUTDIR/$CASE/ehdn/$CASE.manifest.tsv" \
-    --multisample-profile "$OUTPUTDIR/$CASE/ehdn/$CASE.multisample_profile.json" \
-    --output "$OUTPUTDIR/$CASE/ehdn/$CASE.outlier_locus.tsv"
+    --manifest "$OUTPUTDIR/$CASE/str/ehdn/$CASE.manifest.tsv" \
+    --multisample-profile "$OUTPUTDIR/$CASE/str/ehdn/$CASE.multisample_profile.json" \
+    --output "$OUTPUTDIR/$CASE/str/ehdn/$CASE.outlier_locus.tsv"
 
 ehdn_outlier_exitcode=$?
 
@@ -80,9 +92,9 @@ fi
 
 # Run motif_based comparison analysis
 "$EHDN_OUTLIER" motif \
-    --manifest "$OUTPUTDIR/$CASE/ehdn/$CASE.manifest.tsv" \
-    --multisample-profile "$OUTPUTDIR/$CASE/ehdn/$CASE.multisample_profile.json" \
-    --output "$OUTPUTDIR/$CASE/ehdn/$CASE.outlier_motif.tsv"
+    --manifest "$OUTPUTDIR/$CASE/str/ehdn/$CASE.manifest.tsv" \
+    --multisample-profile "$OUTPUTDIR/$CASE/str/ehdn/$CASE.multisample_profile.json" \
+    --output "$OUTPUTDIR/$CASE/str/ehdn/$CASE.outlier_motif.tsv"
 
 ehdn_outlier_exitcode=$?