added container download

Snakefile

@@ -9,6 +9,7 @@ localrules: impute, clean, download_container, generate_samples_file
 # containing all imputed expression values
 rule impute:
     input:
+        "container.sif",
         expand("{output_dir}/gene_expression/{region}.expression.txt", region = config['brain_regions'], output_dir = config['output_dir'])
     output:
         expand("{output_dir}/gene_expressions_combined.rds", output_dir = config['output_dir'])
@@ -30,14 +31,36 @@ rule clean:
 
 # download the required singularity container from zenodo.org
 rule download_container:
+    output:
+        "container_unchecked.sif"
+    shell:
+        """
+        wget https://zenodo.org/record/{config[zenodo_record_id]}/files/container.sif -O container_unchecked.sif
+        """
+
+# check md5
+rule check_container:
+    input:
+        "container_unchecked.sif"
+    output:
+        "container.sif"
+    singularity:
+        "container_unchecked.sif"
     shell:
         """
-        echo 'not implemented'
+        set -e
+        checksum=($(md5sum container_unchecked.sif))
+        if [ $checksum != {config[container_mdf5]} ]; then
+            echo md5 mismatch for container, download corrupted!
+            exit 1
+        fi
+        mv container_unchecked.sif container.sif
         """
 
 # extract dosages in custom PrediXcan format from vcf file
 rule vcf_to_dosages:
     input:
+        "container.sif",
         config['input_vcf_gz_file']
     output:
         "{output_dir}/dosages/chr{i}.dosage.txt.gz"
@@ -63,6 +86,7 @@ rule vcf_to_dosages:
 # extract sample file for PrediXcan
 rule generate_samples_file:
     input:
+        "container.sif",
         config['input_vcf_gz_file']
     output:
         "{output_dir}/dosages/samples.txt"
@@ -83,6 +107,7 @@ rule generate_samples_file:
 # run PrediXcan to impute gene expression for individual tissue type
 rule impute_gene_expression:
     input:
+        "container.sif",
         samples_file = expand("{output_dir}/dosages/samples.txt", output_dir = config['output_dir']),
         dosage_files = expand("{output_dir}/dosages/chr{i}.dosage.txt.gz", i = range(1, 24), output_dir = config['output_dir'])
     output:

config.yml

@@ -23,3 +23,6 @@ brain_regions:
     - 'Putamen_basal_ganglia'
     - 'Spinal_cord_cervical_c-1'
     - 'Substantia_nigra'
+
+zenodo_record_id: '3376504'
+container_mdf5: '1f5aada82c1aeb8e00b0328421ed009b'