Merge pull request #6 from microbiomedata/develop

fixed the WorkflowFailedState caused by set-pipefail and empty results

Author: poeli-lanl

ReadbasedAnalysisTasks.wdl

@@ -10,6 +10,7 @@ task profilerGottcha2 {
     command <<<
         set -euo pipefail
         mkdir -p ${OUTPATH}
+        touch ${OUTPATH}/${PREFIX}.full.tsv
 
         gottcha2.py -r ${RELABD_COL} \
                     -i ${sep=' ' READS} \
@@ -18,7 +19,7 @@ task profilerGottcha2 {
                     -p ${PREFIX} \
                     --database ${DB}
         
-        grep "^species" ${OUTPATH}/${PREFIX}.tsv | ktImportTaxonomy -t 3 -m 9 -o ${OUTPATH}/${PREFIX}.krona.html -
+        grep "^species" ${OUTPATH}/${PREFIX}.tsv | ktImportTaxonomy -t 3 -m 9 -o ${OUTPATH}/${PREFIX}.krona.html - || true
     >>>
     output {
         Map[String, String] results = {

outputTsv2json.py

@@ -59,15 +59,23 @@ def reduceDf(df, cols, ranks=['species','genus','family'], top=10):
 
         # parsing results
         if tool == "gottcha2":
-            df = pd.read_csv(infile, sep='\t')
+            try:
+                df = pd.read_csv(infile, sep='\t')
+            except:
+                pass
+
             if len(df)>0:
                 result['rawResults'] = df.set_index('TAXID').to_dict('split')
                 result['classifiedReadCount'] = df[df['LEVEL']=='superkingdom'].READ_COUNT.sum()
                 result['speciesReadCount'] = df[df['LEVEL']=='species'].READ_COUNT.sum()
                 result['speciesCount'] = len(df[df['LEVEL']=='species'].index)
                 result['taxonomyTop10'] = reduceDf(df, ['LEVEL', 'NAME', 'READ_COUNT', 'REL_ABUNDANCE', 'TAXID'])
         elif tool == "centrifuge":
-            df = pd.read_csv(infile, sep='\t')
+            try:
+                df = pd.read_csv(infile, sep='\t')
+            except:
+                pass
+
             if len(df)>0:
                 df['abundance'] = df['abundance'].astype(float)
                 df['abundance'] = df['abundance']/100
@@ -77,9 +85,13 @@ def reduceDf(df, cols, ranks=['species','genus','family'], top=10):
                 result['speciesCount'] = len(df[df['taxRank']=='species'].index)
                 result['taxonomyTop10'] = reduceDf(df, ['taxRank', 'name', 'numReads', 'abundance', 'taxID'])
         elif tool == "kraken2":
-            df = pd.read_csv(infile,
+            try:
+                df = pd.read_csv(infile,
                             sep='\t', 
                             names=['abundance','numReads','numUniqueReads','taxRank','taxID','name'])
+            except:
+                pass
+
             if len(df)>0:
                 df['abundance'] = df['abundance'].astype(float)
                 df['abundance'] = df['abundance']/100