Filter replicates for only mRNA

scripts/split_study.py → scripts/extract_RIDs.py

@@ -8,7 +8,7 @@ warnings.simplefilter(action='ignore', category=FutureWarning)
 
 def get_args():
     parser = argparse.ArgumentParser()
-    parser.add_argument('-s', '--studyRID',
+    parser.add_argument('-f', '--file',
                         help="The study RID.", required=True)
     args = parser.parse_args()
     return args
@@ -16,13 +16,13 @@ def get_args():
 
 def main():
     args = get_args()
-    studyRID = pd.read_json(args.studyRID+"_studyRID.json")
-    if studyRID["RID"].count() > 0:
-        studyRID["RID"].to_csv(
-            args.studyRID+"_studyRID.csv", header=False, index=False)
+    RID = pd.read_json(args.file+".json")
+    if RID["RID"].count() > 0:
+        RID["RID"].to_csv(
+            args.file+".csv", header=False, index=False, mode='a')
     else:
         raise Exception("No associated replicates found: %s" %
-                        studyRID)
+                        RID)
 
 
 if __name__ == '__main__':

scripts/split_study.sh

@@ -37,19 +37,33 @@ main(){
     shift $(($OPTIND -1))
 
     # Check for mandatory options
-    if [[ -z $rid ]] || [[ -z $tag ]] || [[ -z $email ]] || [[ -z $out ]]; then
+    if [[ -z ${rid} ]] || [[ -z ${tag} ]] || [[ -z ${email} ]] || [[ -z ${out} ]]; then
         usage
     fi
 
-    # query GUDMAP/RBK for study RID
-    echo "curl --location --request GET 'https://www.gudmap.org/ermrest/catalog/2/entity/RNASeq:Replicate/Study_RID="${rid}"'" | bash > $rid\_studyRID.json
-
-    # extract replicate RIDs
-
-    python3 ./scripts/split_study.py -s $rid
+    # query GUDMAP/RBK study RID for experiment RIDs
+    echo "curl --location --request GET 'https://www.gudmap.org/ermrest/catalog/2/entity/RNASeq:Experiment/Sequencing_Type=mRNA-Seq/Study_RID="${rid}"'" | bash > ${rid}\_experimentRIDs.json
+    # extract experiment RIDs
+    >${rid}\_experimentRIDs.csv
+    python3 ./scripts/extract_RIDs.py -f ${rid}\_experimentRIDs
+    expRIDs=$(cat ${rid}\_experimentRIDs.csv)
+
+    > ${rid}\_studyRID.csv
+    for expRID in ${expRIDs}
+    do
+      # query GUDMAP/RBK experiment RIDs for replicate RIDs
+      > temp\_${expRID}\_repRIDs.json
+      echo "curl --location --request GET 'https://www.gudmap.org/ermrest/catalog/2/entity/RNASeq:Replicate/Experiment_RID="${expRID}"'" | bash >> temp\_${expRID}\_repRIDs.json
+      # extract replicate RIDs
+      > temp\_${expRID}\_repRIDs.csv
+      python3 ./scripts/extract_RIDs.py -f temp\_${expRID}\_repRIDs
+      cat temp\_${expRID}\_repRIDs.csv >> ${rid}\_studyRID.csv
+    done
 
     # cleanup study RID files
-    rm $rid\_studyRID.json
+    rm -f temp*
+    rm ${rid}\_experimentRIDs.json
+    rm ${rid}\_experimentRIDs.csv
 
     # make dir for study
     mkdir $out/$rid