added support for fusion novelty type in the abundance utility

src/talon/post/create_abundance_file_from_database.py

@@ -297,6 +297,8 @@ def get_gene_and_transcript_novelty_types(gene_ID, transcript_ID, novelty_type):
     # Look for gene type
     if gene_ID in novelty_type.antisense_genes:
         curr_novel["gene_novelty"] = "Antisense"
+    elif gene_ID in novelty_type.fusion_genes:
+        curr_novel["gene_novelty"] = 'Fusion'
     elif gene_ID in novelty_type.intergenic_genes:
         curr_novel["gene_novelty"] = "Intergenic"
     elif gene_ID in novelty_type.known_genes:
@@ -317,6 +319,8 @@ def get_gene_and_transcript_novelty_types(gene_ID, transcript_ID, novelty_type):
         curr_novel["transcript_novelty"] = "Intergenic"
     elif transcript_ID in novelty_type.genomic_transcripts:
         curr_novel["transcript_novelty"] = "Genomic"
+    elif transcript_ID in novelty_type.fusion_transcripts:
+        curr_novel["transcript_novelty"] = "Fusion"
     elif transcript_ID in novelty_type.known_transcripts:
         curr_novel["transcript_novelty"] = "Known"
     else:
@@ -396,9 +400,14 @@ def make_novelty_type_struct(database, datasets):
     cursor = conn.cursor()
 
     novelty_type = dstruct.Struct()
+
+    # genes
     novelty_type.known_genes = set(qutils.fetch_all_known_genes_detected(cursor, datasets))
     novelty_type.antisense_genes = set(qutils.fetch_antisense_genes(cursor, datasets))
     novelty_type.intergenic_genes = set(qutils.fetch_intergenic_novel_genes(cursor, datasets))
+    novelty_type.fusion_genes = set(qutils.fetch_fusion_novel_genes(cursor, datasets))
+
+    # transcripts
     novelty_type.known_transcripts = set(qutils.fetch_all_known_transcripts_detected(cursor, datasets))
     novelty_type.ISM_transcripts = set(qutils.fetch_all_ISM_transcripts(cursor, datasets))
     novelty_type.ISM_prefix = set(qutils.fetch_prefix_ISM_transcripts(cursor, datasets))
@@ -408,6 +417,7 @@ def make_novelty_type_struct(database, datasets):
     novelty_type.antisense_transcripts = set(qutils.fetch_antisense_transcripts(cursor, datasets))
     novelty_type.intergenic_transcripts = set(qutils.fetch_intergenic_transcripts(cursor, datasets))
     novelty_type.genomic_transcripts = set(qutils.fetch_genomic_transcripts(cursor, datasets))
+    novelty_type.fusion_transcripts = set(qutils.fetch_fusion_transcripts(cursor, datasets))
 
     conn.close()
     return novelty_type

src/talon/query_utils.py

@@ -62,8 +62,8 @@ def fetch_reproducible_NNCs(cursor, datasets):
 
     datasets = format_for_IN(datasets)
     query = (
-        """SELECT gene_ID, 
-                      a.transcript_ID 
+        """SELECT gene_ID,
+                      a.transcript_ID
                FROM abundance as a
 	       LEFT JOIN transcript_annotations as ta
 	           ON ta.ID = a.transcript_ID
@@ -353,14 +353,29 @@ def fetch_intergenic_novel_genes(cursor, datasets):
     genes = [x[0] for x in cursor.fetchall()]
     return genes
 
+def fetch_fusion_novel_genes(cursor, datasets):
+    """Fetch IDs of novel genes denoted as fusion"""
+
+    datasets = format_for_IN(datasets)
+    query = (
+        """SELECT DISTINCT(gene_ID) FROM observed
+                   LEFT JOIN gene_annotations AS ga ON ga.ID = observed.gene_ID
+                   WHERE (ga.attribute = 'fusion_novel')
+                   AND observed.dataset IN """
+        + datasets
+    )
+    cursor.execute(query)
+    genes = [x[0] for x in cursor.fetchall()]
+    return genes
+
 
 def fetch_all_ISM_transcripts(cursor, datasets):
     """Fetch IDs of all ISM transcripts"""
 
     datasets = format_for_IN(datasets)
     query = (
         """SELECT DISTINCT(transcript_ID) FROM observed
-                   LEFT JOIN transcript_annotations 
+                   LEFT JOIN transcript_annotations
                        AS ta ON ta.ID = observed.transcript_ID
                    WHERE (ta.attribute = 'ISM_transcript')
                    AND observed.dataset IN """
@@ -489,6 +504,21 @@ def fetch_genomic_transcripts(cursor, datasets):
     transcripts = [x[0] for x in cursor.fetchall()]
     return transcripts
 
+def fetch_fusion_transcripts(cursor, datasets):
+    """Fetch IDs of all fusion transcripts"""
+
+    datasets = format_for_IN(datasets)
+    query = (
+        """SELECT DISTINCT(transcript_ID) FROM observed
+                   LEFT JOIN transcript_annotations
+                       AS ta ON ta.ID = observed.transcript_ID
+                   WHERE (ta.attribute = 'fusion_transcript')
+                   AND observed.dataset IN """
+        + datasets
+    )
+    cursor.execute(query)
+    transcripts = [x[0] for x in cursor.fetchall()]
+    return transcripts
 
 def fetch_all_transcript_gene_pairs(cursor):
     """Return gene_ID - transcript_ID tuples from database"""