obi clean: first version

python/obitools3/commands/clean.pyx

@@ -0,0 +1,94 @@
+#cython: language_level=3
+
+from obitools3.apps.progress cimport ProgressBar  # @UnresolvedImport
+from obitools3.dms.dms cimport DMS
+from obitools3.dms.capi.obidms cimport OBIDMS_p
+from obitools3.dms.view import RollbackException
+from obitools3.dms.capi.obiclean cimport obi_clean
+from obitools3.apps.optiongroups import addSequenceInputOption, addMinimalOutputOption
+from obitools3.uri.decode import open_uri
+from obitools3.apps.config import logger
+from obitools3.utils cimport tobytes
+
+from obitools3.dms.view.typed_view.view_NUC_SEQS cimport View_NUC_SEQS
+
+
+__title__="Tag a set of sequences for PCR and sequencing errors identification"
+
+
+def addOptions(parser):
+
+    addSequenceInputOption(parser)
+    addMinimalOutputOption(parser)
+ 
+    group = parser.add_argument_group('obi clean specific options')
+
+    group.add_argument('--distance', '-d',
+                       action="store", dest="clean:distance",
+                       metavar='<DISTANCE>',
+                       default=1.0,
+                       type=float,
+                       help="Maximum numbers of errors between two variant sequences. Default: 1.")
+
+    group.add_argument('--sample-tag', '-s',
+                       action="store", 
+                       dest="clean:sample-tag-name",
+                       metavar="<SAMPLE TAG NAME>",
+                       type=str,
+                       default="merged_sample",
+                       help="Name of the tag where sample counts are kept.")
+    
+    group.add_argument('--ratio', '-r',
+                       action="store", dest="clean:ratio",
+                       metavar='<RATIO>',
+                       default=0.5,
+                       type=float,
+                       help="Maximum ratio between the counts of two sequences so that the less abundant one can be considered"
+                            " a variant of the more abundant one. Default: 0.5.")
+ 
+    group.add_argument('--heads-only', '-H',
+                       action="store_true", 
+                       dest="clean:heads-only",
+                       default=False,
+                       help="Only sequences labeled as heads are kept in the output. Default: False")
+
+    group.add_argument('--cluster-tags', '-C',
+                       action="store_true", 
+                       dest="clean:cluster-tags",
+                       default=False,
+                       help="Adds tags for each sequence giving its cluster's head and weight for each sample.")
+
+
+def run(config):
+        
+    DMS.obi_atexit()
+    
+    logger("info", "obi clean")
+    
+    # Open DMS
+    dms_name = config['obi']['inputURI'].split('/')[0]
+    dms = open_uri(dms_name)[0]
+    
+    # Read the name of the input view
+    uri_i = config['obi']['inputURI'].split('/')
+    i_view_name = uri_i[1]
+
+    # Read the name of the output view
+    uri_o = config['obi']['outputURI'].split('/')
+    if len(uri_o)==2:
+        # Check that input and output DMS are the same (predicate, to discuss)
+        if dms_name != uri_o[0]:
+            raise Exception("Input and output DMS must be the same")
+        o_view_name = uri_o[1]
+    else:
+        o_view_name = uri_o[0]
+    
+    if obi_clean(tobytes(dms_name), tobytes(i_view_name), tobytes(config['clean']['sample-tag-name']), tobytes(o_view_name), b"obiclean", \
+              config['clean']['distance'], config['clean']['ratio'], config['clean']['heads-only'], 1) < 0:
+        raise Exception("Error running obiclean")
+
+    print("\n")
+    print(repr(dms[o_view_name]))
+
+    dms.close()
+