Merge pull request #43 from broadinstitute/dp-ksummary

docker build fix and filter_taxids_to_focal_hits fix

Author: dpark01

Dockerfile

@@ -7,7 +7,7 @@ ENV VIRAL_CLASSIFY_PATH=$INSTALL_PATH/viral-classify \
 
 COPY requirements-conda.txt requirements-conda-env2.txt $VIRAL_CLASSIFY_PATH/
 # install most dependencies to the main environment
-RUN $VIRAL_NGS_PATH/docker/install-conda-dependencies.sh $VIRAL_CLASSIFY_PATH/requirements-conda.txt 
+RUN $VIRAL_NGS_PATH/docker/install-conda-dependencies.sh $VIRAL_CLASSIFY_PATH/requirements-conda.txt $VIRAL_NGS_PATH/requirements-conda.txt
 
 # install packages with dependency incompatibilities to the second environment
 RUN CONDA_PREFIX="$MINICONDA_PATH/envs/env2"; \

metagenomics.py

@@ -728,18 +728,16 @@ def filter_taxids_to_focal_hits(taxids_tsv, focal_report_tsv, taxdb_dir, min_rea
     with util.file.open_or_gzopen(focal_report_tsv, "rt") as inf:
         for row in csv.DictReader(inf, delimiter='\t'):
             if int(row['reads_excl_children']) >= min_read_count:
-                hits.add(row['taxon_id'])
+                hits.add(int(row['taxon_id']))
 
     # filter taxids_tsv -> output_tsv
     with util.file.open_or_gzopen(taxids_tsv, "rt") as inf:
         with util.file.open_or_gzopen(output_tsv, "wt") as outf:
             for line in inf:
-                taxid = line.rstrip('\r\n').split('\t')[0]
+                taxid = int(line.rstrip('\r\n').split('\t')[0])
                 ancestors = taxdb.get_ordered_ancestors(taxid)
-                for node in [taxid] + ancestors:
-                    if taxid in hits:
-                        outf.write(line)
-                        break
+                if any(node in hits for node in [taxid] + ancestors):
+                    outf.write(line)
 
 __commands__.append(('filter_taxids_to_focal_hits', parser_filter_taxids_to_focal_hits))
 

requirements-conda.txt

@@ -1,7 +1,7 @@
-blast=2.9.0
+blast>=2.9.0
 bmtagger>=3.101
-diamond>=2.1.6
-kmc>=3.1.1rc1
-kraken2>=2.1.2
+diamond>=2.1.9
+kmc>=3.2.1
+kraken2>=2.1.3
 krona>=2.8.1
-last>=876
+last>=1541

test/unit/test_taxon_filter.py

@@ -4,6 +4,7 @@
     + "[email protected]"
 
 import unittest
+import glob
 import os, os.path
 import tempfile
 import shutil
@@ -364,13 +365,11 @@ def setUp(self):
 
         # tar one db, but not the other
         tar_db_tgz = util.file.mkstempfname('-humanChr9Subset.blastn.db.tar.gz')
-        cmd = ['tar', '-C', self.tempDir, '-cvzf', tar_db_tgz]
-        for ext in ('nhr', 'nin', 'nsq'):
-            cmd.append('humanChr9Subset.'+ext)
+        cmd = ['tar', '-C', self.tempDir, '-cvzf', tar_db_tgz] + list(os.path.basename(f) for f in glob.glob(os.path.join(self.tempDir, "humanChr9Subset.n*")))
         subprocess.check_call(cmd)
         self.blastdbs_multi[1] = tar_db_tgz
-        for ext in ('nhr', 'nin', 'nsq'):
-            os.unlink(os.path.join(self.tempDir, 'humanChr9Subset.'+ext))
+        for idx in glob.glob(os.path.join(self.tempDir, "humanChr9Subset.n*")):
+            os.unlink(idx)
 
     def test_deplete_blastn_bam(self):
         tempDir = tempfile.mkdtemp()