Merge branch 'SRP287614' of https://github.com/metavannier/scRNAseq_analysis_workflow into SRP287614

Author: metavannier

.gitignore

@@ -4,3 +4,12 @@
 02_Container/cellranger.sif
 05_Output
 .vscode/
+.bash_history/
+.cache/
+.conda/
+.config/
+.java/
+.local/
+.singularity/
+snakemake_virtenv/
+

04_Workflow/cellranger.smk

@@ -49,7 +49,7 @@ rule cellranger:
         --jobmode=local \
         --localcores={config[cellranger][localcores]} \
         --localmem={config[cellranger][localmem]}
-        mv ${{sample[$i]}}/out 05_Output/01_cellranger/${{sample[$i]}}/
+        mv ${{sample[$i]}}/outs/* 05_Output/01_cellranger/${{sample[$i]}}/outs/
         rm -r ${{sample[$i]}}/
         mv 05_Output/01_cellranger/${{sample[$i]}}/outs/web_summary.html 05_Output/01_cellranger/${{sample[$i]}}/outs/${{sample[$i]}}_web_summary.html
         done
@@ -93,4 +93,4 @@ rule cellrangeraggr:
         mv ${{outfolder}}/outs/count/* ${{outaggr}}${{outfolder}}/outs/
         mv ${{outfolder}}/outs/web_summary.html ${{outaggr}}/${{outfolder}}/outs/aggregate_web_summary.html
         rm -r ${{outfolder}}
-        """
+        """

04_Workflow/create_snakemake_virtualenv.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+module load userspace/all
+module load python3/3.6.3
+#module load singularity/3.5.1
+
+#mkdir snakemake_virtenv
+virtualenv-3.6 snakemake_virtenv
+source snakemake_virtenv/bin/activate
+pip3 install snakemake==5.22.1
+install numpy deactivate
+#virtualenv-3.6 -p python3 snakemake_virtenv
+

README.md

@@ -48,7 +48,7 @@ Configure the workflow according to your needs via editing the files and reposit
 
 - You need [Singularity v3.5.3](https://github.com/hpcng/singularity/blob/master/INSTALL.md#install-golang) installed on your computer or cluster.
 
-- Load snakemake from a docker container and run the workflow from the root by using these commands:
+- Load snakemake from a docker container and run the workflow from the working directory by using these commands:
 
 `singularity run docker://snakemake/snakemake:v6.3.0`
 
@@ -58,11 +58,11 @@ Configure the workflow according to your needs via editing the files and reposit
 
 #### On a cluster
 
-- Write the batch script to run your snakemake
+- Write the batch script to run your snakemake from the working directory
 
-` sbatch sc_rnaseq_slurm_skylake.sh`
+It will create a snakemake virtual environment and install the packages needed with pip.
 
-singularity exec -H /scratch/tvannier/sc-rnaseq/ docker://snakemake/snakemake:v6.3.0 snakemake --use-conda --use-singularity --cores 12
+`sbatch sc_rnaseq_slurm_skylake.sh`
 
 ### Step 4: Investigate results
 

cluster_config.json

@@ -0,0 +1,10 @@
+__default__:
+    job-name: Job_scrna
+    project: a272
+    partition: skylake
+    time: 09:00:00 # time limit for each job
+    nodes-number: 1
+    cores: 12
+    mem-per-cpu: 5000
+    output: ./%N.%x.out
+    error: ./%N.%x.err

sc_rnaseq_slurm_skylake.sh

@@ -0,0 +1,67 @@
+#!/bin/sh
+#SBATCH -J Job_scrna
+#SBATCH -p skylake
+#SBATCH -N 1
+#SBATCH -n 1
+#SBATCH -A a272
+#SBATCH -t 00:30:00
+#SBATCH -o ./%N.%x.out
+#SBATCH -e ./%N.%x.err
+
+# This script needs to be started from
+# the run directory
+
+# Load the modules and start the virtual environment
+module load userspace/all
+module load python3/3.6.3
+module load singularity/3.5.1
+
+# Create a snakemake virtual environment if necessary
+if [ ! -d snakemake_virtenv/ ]
+then
+  bash 04_Workflow/create_snakemake_virtualenv.sh
+fi
+
+source /scratch/tvannier/sc-rnaseq/snakemake_virtenv/bin/activate
+
+#export PATH=/scratch/tvannier/sc-rnaseq/snakemake_virtenv/condabin:$PATH
+#export LD_LIBRARY_PATH=/scratch/tvannier/sc-rnaseq/snakemake_virtenv/condabin:$LD_LIBRARY_PATH
+
+pip install snakemake==6.3.0
+#pip install manager
+pip install pandas
+#pip install mamba
+
+# move to the working directory
+#cd /scratch/$SLURM_JOB_USER/sc-rnaseq/
+
+
+# ================================================
+# Run the workflow
+# ================================================
+
+# Run snakemake
+snakemake --snakefile Snakefile \
+	--reason \
+	--use-singularity \
+	--use-conda \
+	--conda-frontend conda \
+	--singularity-args="-B /scratch/$SLURM_JOB_USER/sc-rnaseq/" \
+	--jobs 1 \
+	--latency-wait 20 \
+	--max-jobs-per-second 5 \
+	--max-status-checks-per-second 5 \
+	--cluster-config cluster_config.json \
+	--cluster 'sbatch -A {cluster.project} \
+		--job-name {cluster.job-name} \
+		--partition {cluster.partition} \
+		--time {cluster.time} \
+		-N {cluster.nodes-number} \
+		-n {cluster.cores} \
+		--mem-per-cpu {cluster.mem-per-cpu} \
+		--output {cluster.output} \
+		--error {cluster.error}' \
+
+deactivate
+
+