Merge pull request #29 from Merck/develop

Add --prodigal-meta-mode option

Author: prihoda

deepbgc/command/pipeline.py

@@ -59,6 +59,7 @@ def add_arguments(self, parser):
         parser.add_argument('--limit-to-record', action='append', help="Process only specific record ID. Can be provided multiple times")
         parser.add_argument('--minimal-output', dest='is_minimal_output', action='store_true', default=False,
                             help="Produce minimal output with just the GenBank sequence file")
+        parser.add_argument('--prodigal-meta-mode', action='store_true', default=False, help="Run Prodigal in '-p meta' mode to enable detecting genes in short contigs")
         group = parser.add_argument_group('BGC detection options', '')
         no_models_message = 'run "deepbgc download" to download models'
         detector_names = util.get_available_models('detector')
@@ -88,7 +89,7 @@ def add_arguments(self, parser):
 
     def run(self, inputs, output, detectors, no_detector, labels, classifiers, no_classifier,
             is_minimal_output, limit_to_record, score, classifier_score, merge_max_protein_gap, merge_max_nucl_gap, min_nucl,
-            min_proteins, min_domains, min_bio_domains):
+            min_proteins, min_domains, min_bio_domains, prodigal_meta_mode):
         if not detectors:
             detectors = ['deepbgc']
         if not classifiers:
@@ -110,7 +111,7 @@ def run(self, inputs, output, detectors, no_detector, labels, classifiers, no_cl
         output_file_name = os.path.basename(os.path.normpath(output))
 
         steps = []
-        steps.append(DeepBGCAnnotator(tmp_dir_path=tmp_path))
+        steps.append(DeepBGCAnnotator(tmp_dir_path=tmp_path, prodigal_meta_mode=prodigal_meta_mode))
         if not no_detector:
             if not labels:
                 labels = [None] * len(detectors)

deepbgc/command/prepare.py

@@ -32,12 +32,13 @@ class PrepareCommand(BaseCommand):
 
     def add_arguments(self, parser):
         parser.add_argument(dest='inputs', nargs='+', help="Input sequence file path(s) (FASTA/GenBank)")
-        group = parser.add_argument_group('required arguments', '')
         parser.add_argument('--limit-to-record', action='append', help="Process only specific record ID. Can be provided multiple times")
+        group = parser.add_argument_group('required arguments', '')
+        group.add_argument('--prodigal-meta-mode', action='store_true', default=False, help="Run Prodigal in '-p meta' mode to enable detecting genes in short contigs")
         group.add_argument('--output-gbk', required=False, help="Output GenBank file path")
         group.add_argument('--output-tsv', required=False, help="Output TSV file path")
 
-    def run(self, inputs, limit_to_record, output_gbk, output_tsv):
+    def run(self, inputs, limit_to_record, output_gbk, output_tsv, prodigal_meta_mode):
         first_output = output_gbk or output_tsv
         if not first_output:
             raise ValueError('Specify at least one of --output-gbk or --output-tsv')
@@ -47,7 +48,7 @@ def run(self, inputs, limit_to_record, output_gbk, output_tsv):
         if not os.path.exists(tmp_dir_path):
             os.mkdir(tmp_dir_path)
 
-        prepare_step = DeepBGCAnnotator(tmp_dir_path=tmp_dir_path)
+        prepare_step = DeepBGCAnnotator(tmp_dir_path=tmp_dir_path, prodigal_meta_mode=prodigal_meta_mode)
 
         writers = []
         if output_gbk:

deepbgc/pipeline/annotator.py

@@ -8,8 +8,9 @@
 
 class DeepBGCAnnotator(PipelineStep):
 
-    def __init__(self, tmp_dir_path):
+    def __init__(self, tmp_dir_path, prodigal_meta_mode=False):
         self.tmp_dir_path = tmp_dir_path
+        self.prodigal_meta_mode = prodigal_meta_mode
 
     def run(self, record):
         logging.info('Preparing record %s', record.id)
@@ -25,7 +26,7 @@ def run(self, record):
         if num_proteins:
             logging.info('Sequence already contains %s CDS features, skipping CDS detection', num_proteins)
         else:
-            protein_annotator = ProdigalProteinRecordAnnotator(record=record, tmp_path_prefix=record_tmp_path)
+            protein_annotator = ProdigalProteinRecordAnnotator(record=record, tmp_path_prefix=record_tmp_path, meta_mode=self.prodigal_meta_mode)
             protein_annotator.annotate()
 
         num_pfams = len(util.get_pfam_features(record))

deepbgc/pipeline/protein.py

@@ -12,9 +12,10 @@
 
 
 class ProdigalProteinRecordAnnotator(object):
-    def __init__(self, record, tmp_path_prefix):
+    def __init__(self, record, tmp_path_prefix, meta_mode=False):
         self.record = record
         self.tmp_path_prefix = tmp_path_prefix
+        self.meta_mode = meta_mode
 
     def annotate(self):
         logging.info('Finding genes in record: %s', self.record.id)
@@ -29,8 +30,10 @@ def annotate(self):
 
         logging.debug('Detecting genes using Prodigal...')
 
+
+
         p = subprocess.Popen(
-            ['prodigal', '-i', nucl_path, '-a', protein_path],
+            ['prodigal', '-i', nucl_path, '-a', protein_path] + (['-p','meta'] if self.meta_mode else []),
             stdout=subprocess.PIPE,
             stderr=subprocess.PIPE,
             universal_newlines=True
@@ -42,7 +45,7 @@ def annotate(self):
             logging.warning('== End Prodigal Error. ============')
 
             if 'Sequence must be' in err:
-                logging.warning('No proteins detected in short sequence, moving on.')
+                logging.warning('No proteins detected in short sequence, use --prodigal-meta-mode to run Prodigal in "-p meta" mode.')
             elif os.stat(protein_path).st_size == 0:
                 raise ValueError("Prodigal produced empty output, make sure to use a DNA sequence.")
             else: