Check GenBank and RefSeq for data to download

bin/parse_preview_json.py

@@ -0,0 +1,14 @@
+#!/usr/bin/env python3
+"""Parse the preview.json produced by the NCBI datasets CLI"""
+
+import json
+
+dat = json.load(open('preview.json'))
+print("Checking for number of records")
+print(json.dumps(dat, indent=4))
+
+for kw, val in dat["included_data_files"].items():
+    if val["file_count"] > 0:
+        with open("accession.has.data", "w") as handle:
+            handle.write('TRUE')
+        break

templates/getDatasets.sh

@@ -2,5 +2,34 @@
 
 datasets --version
 
-echo "Downloading accession ${dataset_acc}"
-datasets download genome accession ${dataset_acc} --no-progressbar --include ${params.ncbi_datasets_type}
+hasData(){
+    # Check to see if an accession has data available
+    echo "Checking for data in \$1"
+    datasets \
+        download \
+        genome \
+        accession \
+        \$1 \
+        --no-progressbar \
+        --include ${params.ncbi_datasets_type} \
+        --preview > preview.json
+
+    # Parse the number of records
+    parse_preview_json.py
+    if [ -s accession.has.data ]; then
+        echo "Data found for \$1"
+        return 0
+    else
+        echo "No data found for \$1"
+        return 1
+    fi
+}
+
+INPUT_ACC=${dataset_acc}
+for ACC in \${INPUT_ACC} \${INPUT_ACC/A_/F_} \${INPUT_ACC/F_/A_}; do
+    if hasData \$ACC; then
+        echo "Downloading accession \${ACC}"
+        datasets download genome accession \${ACC} --no-progressbar --include ${params.ncbi_datasets_type}
+        break
+    fi
+done